Layers that will be compiled:
  group_conv_0
  single_conv_0
  group_conv_1
  single_conv_1
  group_pre_2
  group_post_2
  single_pre_2
  single_post_2
  group_conv_3
  single_conv_3
  group_conv_4
  single_conv_4
  group_pre_5
  group_post_5
  single_pre_5
  single_post_5
  group_conv_6
  single_conv_6
  group_conv_7
  single_conv_7
  group_conv_8
  single_conv_8
  group_pre_9
  group_post_9
  single_pre_9
  single_post_9
  group_conv_10
  single_conv_10
  group_conv_11
  single_conv_11
  group_conv_12
  single_conv_12
  group_pre_13
  group_post_13
  single_pre_13
  single_post_13
  group_conv_14
  single_conv_14
  group_conv_15
  single_conv_15
  group_conv_16
  single_conv_16
  group_pre_17
  group_post_17
  single_pre_17
  single_post_17
  group_conv_18
  single_conv_18
  group_conv_19
  single_conv_19
  group_conv_20
  single_conv_20
  group_pre_21
  group_post_21
  single_pre_21
  single_post_21
  group_conv_22
  single_conv_22
  group_conv_23
  single_conv_23
  group_pre_24
  group_post_24
  single_pre_24
  single_post_24
  group_conv_25
  single_conv_25
  group_conv_26
  single_conv_26
  group_pre_27
  group_post_27
  single_pre_27
  single_post_27
  group_conv_28
  single_conv_28
  group_conv_29
  single_conv_29
  group_cache_0
  group_cache_128
  group_cache_256
  group_cache_384
  group_cache_512
  group_cache_640
  group_cache_768
  group_cache_896
  group_cache_1024
  group_cache_1152
  group_cache_1280
  group_cache_1408
  group_cache_1536
  group_cache_1664
  group_cache_1792
  group_cache_1920
  single_cache_127
  single_cache_255
  single_cache_383
  single_cache_511
  single_cache_639
  single_cache_767
  single_cache_895
  single_cache_1023
  single_cache_1151
  single_cache_1279
  single_cache_1407
  single_cache_1535
  single_cache_1663
  single_cache_1791
  single_cache_1919
  single_cache_2047
  conv_post_final_29

2026-03-07 05:31:50,582 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.DEVKIT files...
Generated all mode=DEVKIT files
2026-03-07 05:31:56,037 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.SOURCE_TO_ONNX files...
2026-03-07 05:34:57,569 - sima_lmm.model.vision_language_model - INFO - FileGenMode.SOURCE_TO_ONNX files generation completed.
Generated all mode=SOURCE_TO_ONNX files
2026-03-07 05:34:57,570 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.ONNX_TO_QUANT files...
2026-03-07 05:35:22,842 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv.onnx'] in onnx format
2026-03-07 05:35:22,842 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv.onnx'] in onnx format
2026-03-07 05:35:22,842 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv.onnx'] in onnx format
2026-03-07 05:35:22,843 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2.onnx'] in onnx format
2026-03-07 05:35:22,846 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv.onnx'] in onnx format
2026-03-07 05:35:22,848 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer2.onnx'] in onnx format
2026-03-07 05:35:24,079 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:24,079 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:26,033 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:26,033 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:26,227 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:26,227 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:26,244 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:26,245 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:26,518 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:26,518 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:26,599 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:26,599 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:30,037 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:32,031 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2.sima
2026-03-07 05:35:32,034 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2.onnx'] in onnx format
2026-03-07 05:35:32,576 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:32,616 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:33,202 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:33,203 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:35,884 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:38,141 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:38,334 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:40,245 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:43,676 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2.sima
2026-03-07 05:35:43,680 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer2.onnx'] in onnx format
2026-03-07 05:35:44,252 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:35:44,515 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv.sima
2026-03-07 05:35:44,555 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv.onnx'] in onnx format
2026-03-07 05:35:44,555 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:35:44,615 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer2.sima
2026-03-07 05:35:44,619 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv.onnx'] in onnx format
2026-03-07 05:35:45,332 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv.sima
2026-03-07 05:35:45,373 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv.onnx'] in onnx format
2026-03-07 05:35:46,476 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:46,476 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:46,707 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:46,707 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:47,680 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv.sima
2026-03-07 05:35:47,692 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:47,692 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:47,699 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv.onnx'] in onnx format
2026-03-07 05:35:48,073 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv.sima
2026-03-07 05:35:48,095 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5.onnx'] in onnx format
2026-03-07 05:35:48,658 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:48,659 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:49,107 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:49,107 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:49,864 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:49,864 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:51,242 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:52,563 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:54,926 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:55,887 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:35:57,020 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5.sima
2026-03-07 05:35:57,024 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer5.onnx'] in onnx format
2026-03-07 05:35:57,905 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer2.sima
2026-03-07 05:35:57,944 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5.onnx'] in onnx format
2026-03-07 05:35:59,023 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:59,023 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:59,141 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:35:59,141 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:35:59,497 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:00,156 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:00,704 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv.sima
2026-03-07 05:36:00,718 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer5.onnx'] in onnx format
2026-03-07 05:36:03,389 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:03,389 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:04,565 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:36:05,252 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:36:06,119 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:06,357 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv.sima
2026-03-07 05:36:06,402 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv.onnx'] in onnx format
2026-03-07 05:36:07,510 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5.sima
2026-03-07 05:36:07,523 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv.onnx'] in onnx format
2026-03-07 05:36:08,157 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv.sima
2026-03-07 05:36:08,167 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv.onnx'] in onnx format
2026-03-07 05:36:08,602 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:08,742 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv.sima
2026-03-07 05:36:08,751 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv.onnx'] in onnx format
2026-03-07 05:36:09,132 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:09,132 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:10,477 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:10,477 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:10,946 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:11,241 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:11,242 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:11,549 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:11,549 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:15,023 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer5.sima
2026-03-07 05:36:15,036 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv.onnx'] in onnx format
2026-03-07 05:36:16,527 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer5.sima
2026-03-07 05:36:16,544 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv.onnx'] in onnx format
2026-03-07 05:36:16,796 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:17,552 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:17,552 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:17,671 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:19,377 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:19,377 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:21,355 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:22,748 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:26,613 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:26,826 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv.sima
2026-03-07 05:36:26,897 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9.onnx'] in onnx format
2026-03-07 05:36:27,721 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv.sima
2026-03-07 05:36:27,772 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer9.onnx'] in onnx format
2026-03-07 05:36:28,039 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:28,039 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:29,676 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv.sima
2026-03-07 05:36:29,695 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9.onnx'] in onnx format
2026-03-07 05:36:30,427 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:30,428 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:30,453 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:30,833 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:30,833 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:30,937 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv.sima
2026-03-07 05:36:30,954 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer9.onnx'] in onnx format
2026-03-07 05:36:33,090 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:33,090 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:33,861 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:35,845 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9.sima
2026-03-07 05:36:35,847 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv.sima
2026-03-07 05:36:35,851 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv.onnx'] in onnx format
2026-03-07 05:36:35,873 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv.onnx'] in onnx format
2026-03-07 05:36:36,860 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:37,634 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv.sima
2026-03-07 05:36:37,662 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv.onnx'] in onnx format
2026-03-07 05:36:37,665 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:38,139 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9.sima
2026-03-07 05:36:38,145 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv.onnx'] in onnx format
2026-03-07 05:36:38,759 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:38,759 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:39,069 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:39,069 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:40,247 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:40,248 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:40,827 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:40,827 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:41,006 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:44,226 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer9.sima
2026-03-07 05:36:44,249 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv.onnx'] in onnx format
2026-03-07 05:36:45,415 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:46,619 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer9.sima
2026-03-07 05:36:46,639 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv.onnx'] in onnx format
2026-03-07 05:36:46,876 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:46,876 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:47,341 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:49,597 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:49,597 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:50,632 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:51,930 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:54,375 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv.sima
2026-03-07 05:36:54,402 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13.onnx'] in onnx format
2026-03-07 05:36:55,961 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:36:55,962 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:36:57,516 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv.sima
2026-03-07 05:36:57,529 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer13.onnx'] in onnx format
2026-03-07 05:36:57,809 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:36:59,818 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv.sima
2026-03-07 05:36:59,832 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13.onnx'] in onnx format
2026-03-07 05:37:00,112 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:00,243 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:00,243 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:00,390 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv.sima
2026-03-07 05:37:00,410 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer13.onnx'] in onnx format
2026-03-07 05:37:00,862 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:00,862 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:01,975 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:04,509 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:04,510 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:04,670 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13.sima
2026-03-07 05:37:04,675 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv.onnx'] in onnx format
2026-03-07 05:37:07,351 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:07,351 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:07,978 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:08,116 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv.sima
2026-03-07 05:37:08,171 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv.onnx'] in onnx format
2026-03-07 05:37:08,343 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv.sima
2026-03-07 05:37:08,355 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv.onnx'] in onnx format
2026-03-07 05:37:09,268 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13.sima
2026-03-07 05:37:09,272 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv.onnx'] in onnx format
2026-03-07 05:37:09,443 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:11,106 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:11,106 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:11,289 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:11,470 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:11,470 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:11,649 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:11,649 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:16,736 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer13.sima
2026-03-07 05:37:16,741 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv.onnx'] in onnx format
2026-03-07 05:37:17,267 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer13.sima
2026-03-07 05:37:17,302 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv.onnx'] in onnx format
2026-03-07 05:37:17,438 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:18,053 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:19,943 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:19,999 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:19,999 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:20,551 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:20,551 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:23,366 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:27,203 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv.sima
2026-03-07 05:37:27,227 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17.onnx'] in onnx format
2026-03-07 05:37:27,335 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:27,537 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv.sima
2026-03-07 05:37:27,566 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer17.onnx'] in onnx format
2026-03-07 05:37:28,266 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:28,266 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:28,381 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv.sima
2026-03-07 05:37:28,415 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17.onnx'] in onnx format
2026-03-07 05:37:29,460 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:29,461 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:30,180 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:30,180 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:30,518 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv.sima
2026-03-07 05:37:30,538 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer17.onnx'] in onnx format
2026-03-07 05:37:31,428 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:32,604 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:32,604 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:34,381 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:36,615 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17.sima
2026-03-07 05:37:36,620 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv.onnx'] in onnx format
2026-03-07 05:37:36,914 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:37,355 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv.sima
2026-03-07 05:37:37,413 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv.onnx'] in onnx format
2026-03-07 05:37:37,702 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:38,283 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17.sima
2026-03-07 05:37:38,292 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv.onnx'] in onnx format
2026-03-07 05:37:39,328 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv.sima
2026-03-07 05:37:39,358 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv.onnx'] in onnx format
2026-03-07 05:37:39,459 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:39,459 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:40,288 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:40,288 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:40,836 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:41,035 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:41,035 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:41,088 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:41,088 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:44,198 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer17.sima
2026-03-07 05:37:44,204 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv.onnx'] in onnx format
2026-03-07 05:37:46,301 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer17.sima
2026-03-07 05:37:46,317 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv.onnx'] in onnx format
2026-03-07 05:37:46,838 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:47,141 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:47,217 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:47,217 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:49,429 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:49,430 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:51,036 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:52,533 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:57,664 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:57,740 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv.sima
2026-03-07 05:37:57,746 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21.onnx'] in onnx format
2026-03-07 05:37:58,034 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv.sima
2026-03-07 05:37:58,082 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer21.onnx'] in onnx format
2026-03-07 05:37:58,878 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:37:58,878 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:37:59,342 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:37:59,438 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv.sima
2026-03-07 05:37:59,466 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21.onnx'] in onnx format
2026-03-07 05:38:00,681 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:00,682 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:00,810 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:00,810 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:00,950 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv.sima
2026-03-07 05:38:00,962 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer21.onnx'] in onnx format
2026-03-07 05:38:03,413 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:03,413 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:05,401 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:06,583 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv.sima
2026-03-07 05:38:06,645 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:06,652 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv.onnx'] in onnx format
2026-03-07 05:38:07,354 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv.sima
2026-03-07 05:38:07,380 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv.onnx'] in onnx format
2026-03-07 05:38:07,570 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21.sima
2026-03-07 05:38:07,578 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv.onnx'] in onnx format
2026-03-07 05:38:08,083 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21.sima
2026-03-07 05:38:08,089 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv.onnx'] in onnx format
2026-03-07 05:38:08,114 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:09,194 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:09,194 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:09,824 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:09,824 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:09,920 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:09,921 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:10,941 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:10,941 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:11,126 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:14,993 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:15,276 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer21.sima
2026-03-07 05:38:15,305 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24.onnx'] in onnx format
2026-03-07 05:38:16,339 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:16,340 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:17,307 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer21.sima
2026-03-07 05:38:17,344 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer24.onnx'] in onnx format
2026-03-07 05:38:17,700 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:19,555 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:19,555 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:21,189 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:21,294 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:22,388 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:24,571 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv.sima
2026-03-07 05:38:24,574 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24.onnx'] in onnx format
2026-03-07 05:38:26,538 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24.sima
2026-03-07 05:38:26,541 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer24.onnx'] in onnx format
2026-03-07 05:38:26,557 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:26,561 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:27,790 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:38:27,930 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:38:29,089 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:29,089 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:29,451 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv.sima
2026-03-07 05:38:29,493 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv.onnx'] in onnx format
2026-03-07 05:38:31,202 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv.sima
2026-03-07 05:38:31,213 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv.onnx'] in onnx format
2026-03-07 05:38:31,383 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv.sima
2026-03-07 05:38:31,403 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv.onnx'] in onnx format
2026-03-07 05:38:32,077 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:32,568 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:32,568 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:32,929 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:32,929 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:33,328 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:34,526 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:34,787 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:34,791 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:35,554 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24.sima
2026-03-07 05:38:35,560 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv.onnx'] in onnx format
2026-03-07 05:38:37,203 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:37,204 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:38,418 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer24.sima
2026-03-07 05:38:38,421 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27.onnx'] in onnx format
2026-03-07 05:38:39,421 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:39,421 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:39,608 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:41,196 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer24.sima
2026-03-07 05:38:41,220 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_post_layer27.onnx'] in onnx format
2026-03-07 05:38:43,078 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:43,401 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:43,401 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:44,587 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:45,346 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:46,850 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:48,522 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27.sima
2026-03-07 05:38:48,525 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27.onnx'] in onnx format
2026-03-07 05:38:49,015 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv.sima
2026-03-07 05:38:49,018 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer27.onnx'] in onnx format
2026-03-07 05:38:49,971 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:49,971 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:50,187 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:38:50,831 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:50,831 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:53,624 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv.sima
2026-03-07 05:38:53,630 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv.onnx'] in onnx format
2026-03-07 05:38:53,665 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv.sima
2026-03-07 05:38:53,675 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv.onnx'] in onnx format
2026-03-07 05:38:55,178 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:55,260 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv.sima
2026-03-07 05:38:55,265 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv.onnx'] in onnx format
2026-03-07 05:38:55,659 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:55,660 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:55,852 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:55,852 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:55,972 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:56,019 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:56,019 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:56,445 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:38:58,881 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27.sima
2026-03-07 05:38:58,887 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv.onnx'] in onnx format
2026-03-07 05:38:59,307 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:38:59,307 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:38:59,436 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:01,470 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv.sima
2026-03-07 05:39:01,475 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token0.onnx'] in onnx format
2026-03-07 05:39:01,563 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:01,563 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:02,097 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer27.sima
2026-03-07 05:39:02,114 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token128.onnx'] in onnx format
2026-03-07 05:39:02,212 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:02,213 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:02,373 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer27.sima
2026-03-07 05:39:02,381 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token256.onnx'] in onnx format
2026-03-07 05:39:02,481 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:02,482 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:02,682 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:02,697 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:04,612 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:05,146 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv.sima
2026-03-07 05:39:05,152 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token384.onnx'] in onnx format
2026-03-07 05:39:05,241 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:05,241 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:05,352 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:05,528 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:06,534 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token0.sima
2026-03-07 05:39:06,537 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token512.onnx'] in onnx format
2026-03-07 05:39:06,634 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:06,635 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:07,822 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token128.sima
2026-03-07 05:39:07,824 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token640.onnx'] in onnx format
2026-03-07 05:39:07,910 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:07,911 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:08,427 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:08,580 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token256.sima
2026-03-07 05:39:08,583 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token768.onnx'] in onnx format
2026-03-07 05:39:08,670 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:08,671 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:08,937 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:09,778 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:11,252 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:12,999 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv.sima
2026-03-07 05:39:13,029 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token896.onnx'] in onnx format
2026-03-07 05:39:13,039 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:13,257 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:13,257 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:14,029 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token384.sima
2026-03-07 05:39:14,032 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024.onnx'] in onnx format
2026-03-07 05:39:14,123 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:14,124 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:15,339 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:15,498 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token512.sima
2026-03-07 05:39:15,501 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152.onnx'] in onnx format
2026-03-07 05:39:15,591 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:15,591 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:16,547 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:16,639 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv.sima
2026-03-07 05:39:16,647 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280.onnx'] in onnx format
2026-03-07 05:39:16,770 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:16,824 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:16,824 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:17,037 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token640.sima
2026-03-07 05:39:17,041 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408.onnx'] in onnx format
2026-03-07 05:39:17,131 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:17,132 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:17,488 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:18,582 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:18,648 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:18,845 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token768.sima
2026-03-07 05:39:18,849 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536.onnx'] in onnx format
2026-03-07 05:39:18,941 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:18,941 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:20,405 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:20,454 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:22,323 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:22,981 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:23,141 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token896.sima
2026-03-07 05:39:23,144 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664.onnx'] in onnx format
2026-03-07 05:39:23,236 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:23,237 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:25,564 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:25,733 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024.sima
2026-03-07 05:39:25,736 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792.onnx'] in onnx format
2026-03-07 05:39:25,826 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:25,826 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:26,483 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:26,644 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152.sima
2026-03-07 05:39:26,648 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920.onnx'] in onnx format
2026-03-07 05:39:26,668 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:26,814 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:26,814 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:28,891 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:29,278 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:29,472 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280.sima
2026-03-07 05:39:29,476 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token127.onnx'] in onnx format
2026-03-07 05:39:29,567 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:29,567 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:29,988 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:30,149 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408.sima
2026-03-07 05:39:30,154 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token255.onnx'] in onnx format
2026-03-07 05:39:30,228 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:30,254 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:30,255 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:32,254 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:32,410 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536.sima
2026-03-07 05:39:32,414 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token383.onnx'] in onnx format
2026-03-07 05:39:32,503 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:32,504 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:32,833 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:33,463 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:33,755 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token127.sima
2026-03-07 05:39:33,761 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token511.onnx'] in onnx format
2026-03-07 05:39:33,901 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:33,902 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:34,372 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token255.sima
2026-03-07 05:39:34,375 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token639.onnx'] in onnx format
2026-03-07 05:39:34,463 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:34,463 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:35,604 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:36,643 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:36,798 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token383.sima
2026-03-07 05:39:36,801 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token767.onnx'] in onnx format
2026-03-07 05:39:36,867 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664.sima
2026-03-07 05:39:36,870 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token895.onnx'] in onnx format
2026-03-07 05:39:36,888 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:36,888 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:36,955 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:36,955 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:37,045 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:37,581 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:38,195 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token511.sima
2026-03-07 05:39:38,198 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023.onnx'] in onnx format
2026-03-07 05:39:38,283 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:38,283 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:38,859 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token639.sima
2026-03-07 05:39:38,862 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151.onnx'] in onnx format
2026-03-07 05:39:38,947 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:38,947 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:39,732 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:39,887 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792.sima
2026-03-07 05:39:39,891 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279.onnx'] in onnx format
2026-03-07 05:39:39,976 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:39,977 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:40,024 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:40,103 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:41,387 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:41,759 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token767.sima
2026-03-07 05:39:41,762 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407.onnx'] in onnx format
2026-03-07 05:39:41,849 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:41,849 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:41,935 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token895.sima
2026-03-07 05:39:41,938 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535.onnx'] in onnx format
2026-03-07 05:39:42,025 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:42,025 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:42,095 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples
2026-03-07 05:39:42,252 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920.sima
2026-03-07 05:39:42,255 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663.onnx'] in onnx format
2026-03-07 05:39:42,340 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:42,340 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:42,680 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:42,925 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:43,146 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023.sima
2026-03-07 05:39:43,149 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791.onnx'] in onnx format
2026-03-07 05:39:43,240 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:43,240 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:44,822 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151.sima
2026-03-07 05:39:44,825 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919.onnx'] in onnx format
2026-03-07 05:39:44,911 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:44,911 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:44,997 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:45,039 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:45,140 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279.sima
2026-03-07 05:39:45,143 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047.onnx'] in onnx format
2026-03-07 05:39:45,316 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:45,316 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:45,809 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:47,233 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:47,824 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407.sima
2026-03-07 05:39:47,827 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-2.6B/onnx_files/models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final.onnx'] in onnx format
2026-03-07 05:39:48,161 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535.sima
2026-03-07 05:39:48,321 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663.sima
2026-03-07 05:39:48,452 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:49,055 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:39:49,613 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791.sima
2026-03-07 05:39:51,174 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919.sima
2026-03-07 05:39:51,445 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047.sima
2026-03-07 05:39:52,773 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False
2026-03-07 05:39:52,773 - afe.apis.loaded_net - INFO - Calibration method = mse
2026-03-07 05:39:59,042 - afe.backends.mla.mla_checkers - INFO - Cannot assign node cast_10, source_name(['argmax']) to MLA.
	['Unsupported']
2026-03-07 05:39:59,718 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ...
2026-03-07 05:40:19,244 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final.sima
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Match check_no_dynamic_weights pattern
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
Running Calibration ...

Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete.  1/1
Running Calibration ...DONE
Running quantization ...
Running quantization ...DONE
2026-03-07 05:40:21,673 - sima_lmm.model.vision_language_model - INFO - FileGenMode.ONNX_TO_QUANT files generation completed.
Generated all mode=ONNX_TO_QUANT files
2026-03-07 05:40:21,673 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.MODEL_SDK_COMPILE files...
2026-03-07 05:40:47,213 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv.sima
2026-03-07 05:40:47,301 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:47,301 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv"
2026-03-07 05:40:47,304 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:47,304 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:47,339 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:47,339 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:47,339 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:47,339 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:47,385 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv.sima
2026-03-07 05:40:47,393 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2.sima
2026-03-07 05:40:47,413 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:47,418 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:47,418 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2"
2026-03-07 05:40:47,422 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:47,423 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:47,430 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:47,431 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:47,431 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:47,431 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:47,441 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:47,462 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:47,477 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:47,477 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv"
2026-03-07 05:40:47,481 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:47,481 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:47,517 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:47,518 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:47,518 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:47,518 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:47,595 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:47,623 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:47,632 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv.sima
2026-03-07 05:40:47,644 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:47,651 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv.sima
2026-03-07 05:40:47,686 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:47,687 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:47,705 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:47,756 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:47,756 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv"
2026-03-07 05:40:47,762 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:47,762 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:47,770 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:47,770 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv"
2026-03-07 05:40:47,776 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:47,776 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:47,813 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:47,813 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:47,813 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:47,813 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:47,828 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:47,829 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:47,829 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:47,829 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:47,869 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:47,870 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:47,884 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:48,019 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer2.sima
2026-03-07 05:40:48,128 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:40:48,128 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer2"
2026-03-07 05:40:48,133 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:40:48,133 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:40:48,176 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:40:48,177 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:40:48,177 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:40:48,177 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:40:53,931 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:54,293 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:54,361 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:54,365 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:54,401 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:54,413 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:54,444 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:54,509 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:40:54,575 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:54,575 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:54,592 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:54,865 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:54,872 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:54,873 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:54,889 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:54,930 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:54,942 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:40:55,009 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:40:55,304 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:55,305 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:55,321 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:55,383 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:40:55,384 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:40:55,401 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:40:55,728 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:40:55,771 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:40:55,958 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:40:56,009 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:00,142 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:00,393 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:01,276 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:01,500 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:01,544 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:01,562 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:01,792 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:01,834 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:04,888 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:05,021 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:07,092 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:07,212 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:10,084 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:10,199 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:10,232 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:10,350 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:10,381 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.975
2026-03-07 05:41:10,382 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:10,622 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:10,704 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.969
2026-03-07 05:41:10,705 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:10,755 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:10,755 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:10,946 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:11,080 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:11,080 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:16,054 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:17,099 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:17,614 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:17,661 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:17,983 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:18,983 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.95
2026-03-07 05:41:18,983 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:19,008 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:19,157 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:19,157 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:19,171 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:19,746 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:19,838 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:23,067 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:23,462 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:24,546 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:24,948 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:25,127 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:41:25,135 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:41:25,242 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:25,349 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:25,646 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:25,752 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:41:29,705 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:29,706 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer2_mpk.json, llima-compile
2026-03-07 05:41:30,107 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2.sima
2026-03-07 05:41:30,130 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:41:30,130 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2"
2026-03-07 05:41:30,134 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:41:30,134 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:41:30,139 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:41:30,139 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:41:30,139 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:41:30,139 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:41:30,311 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:41:30,331 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:41:30,364 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:41:30,392 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:41:30,393 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:41:30,402 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:41:32,897 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:32,927 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:34,196 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.971
2026-03-07 05:41:34,197 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:34,377 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:34,425 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:41:34,438 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:41:34,674 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:41:34,688 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:34,688 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:34,688 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:41:34,967 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:35,322 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:35,436 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:35,448 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:36,163 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.974
2026-03-07 05:41:36,164 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:36,195 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:36,234 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:36,234 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:38,563 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,025 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer1_conv_stage1_mla.elf, llima-compile
2026-03-07 05:41:40,205 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer2.sima
2026-03-07 05:41:40,275 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:41:40,275 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer2"
2026-03-07 05:41:40,277 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:41:40,277 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:41:40,305 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:41:40,306 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:41:40,306 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:41:40,306 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:40,308 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer0_conv_mpk.json, llima-compile
2026-03-07 05:41:40,346 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:41:40,351 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:41:40,362 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:41:40,545 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:41:40,545 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:41:40,554 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:41:41,021 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv.sima
2026-03-07 05:41:41,127 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:41:41,127 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv"
2026-03-07 05:41:41,132 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:41:41,132 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:41:41,185 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:41:41,185 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:41:41,185 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:41:41,186 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:41:42,441 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.955
2026-03-07 05:41:42,441 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:42,655 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:42,746 - mlc.test_util.test_context - INFO - Compression done in 16s. Compression ratio: 0.963
2026-03-07 05:41:42,746 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:42,961 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:43,009 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:43,009 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:43,327 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:43,327 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:41:47,330 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:41:47,363 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:41:47,963 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:41:48,395 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:41:48,462 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:41:48,831 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:41:48,832 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:41:48,848 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:41:50,338 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:41:51,199 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:41:51,338 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:41:51,376 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:41:51,410 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:41:52,298 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:41:52,298 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:41:52,298 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:52,298 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:41:52,298 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:41:52,299 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer2_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:41:52,800 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv.sima
2026-03-07 05:41:52,884 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:41:52,885 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv"
2026-03-07 05:41:52,887 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:41:52,888 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:41:52,920 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:41:52,921 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:41:52,921 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:41:52,921 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:41:52,995 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:41:53,022 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:41:53,043 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:41:53,262 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:41:53,263 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:41:53,276 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:41:58,829 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.983
2026-03-07 05:41:58,829 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:41:59,029 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:41:59,141 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:41:59,141 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:42:01,860 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:42:01,903 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:42:04,010 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:42:04,163 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:42:05,620 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:42:06,761 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:42:06,987 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:42:07,028 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:42:14,722 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:42:14,736 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 11s.
2026-03-07 05:42:15,795 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.969
2026-03-07 05:42:15,795 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:42:15,988 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:42:16,035 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:42:16,169 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:42:16,169 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:42:17,463 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:42:18,182 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:42:18,293 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:42:19,297 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:42:19,297 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:42:23,878 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:42:23,878 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:42:23,878 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:23,879 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer2_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_post_layer2_stage1_mla.elf, llima-compile
2026-03-07 05:42:24,581 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv.sima
2026-03-07 05:42:24,668 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:42:24,668 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv"
2026-03-07 05:42:24,673 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:42:24,673 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:42:24,722 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:42:24,723 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:42:24,723 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:42:24,723 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:25,520 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer2_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer2_stage1_mla.elf, llima-compile
2026-03-07 05:42:25,823 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv.sima
2026-03-07 05:42:25,912 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:42:25,912 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv"
2026-03-07 05:42:25,915 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:42:25,915 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:42:25,949 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:42:25,950 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:42:25,950 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:42:25,950 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:42:26,024 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:42:26,052 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:42:26,073 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:42:26,294 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:42:26,295 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:42:26,308 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:42:29,938 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:42:29,952 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:42:30,841 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:42:30,855 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:42:31,430 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:42:31,861 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:42:31,929 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:42:32,277 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:42:32,278 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:42:32,295 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:42:34,832 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:42:34,878 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:42:35,854 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.956
2026-03-07 05:42:35,854 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:42:36,080 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:42:36,467 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:42:36,467 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:42:38,720 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:42:39,857 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:42:40,086 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:42:40,128 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:42:40,393 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:42:40,393 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:45,809 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer0_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,046 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer3_conv_stage1_mla.elf, llima-compile
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:42:46,104 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer1_conv_mpk.json, llima-compile
2026-03-07 05:42:46,224 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5.sima
2026-03-07 05:42:46,245 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:42:46,245 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5"
2026-03-07 05:42:46,248 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:42:46,249 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:42:46,254 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:42:46,254 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:42:46,254 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:42:46,254 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:42:46,515 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5.sima
2026-03-07 05:42:46,538 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:42:46,538 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5"
2026-03-07 05:42:46,542 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:42:46,542 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:42:46,547 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:42:46,548 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:42:46,548 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:42:46,548 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:42:46,657 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer5.sima
2026-03-07 05:42:46,720 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:42:46,740 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:42:46,749 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:42:46,749 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer5"
2026-03-07 05:42:46,753 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:42:46,753 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:42:46,774 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:42:46,797 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:42:46,797 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:42:46,797 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:42:46,797 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:42:46,804 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:42:46,804 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:42:46,813 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:42:47,484 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:42:47,636 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:42:48,976 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.973
2026-03-07 05:42:48,977 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:42:49,215 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:42:49,333 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:42:49,352 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:42:49,352 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:42:49,365 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:42:51,465 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:42:51,819 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:42:51,935 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:42:51,947 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:42:52,665 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.969
2026-03-07 05:42:52,665 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:42:52,693 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:42:52,704 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:42:52,734 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:42:52,734 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:42:52,778 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:42:52,832 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:42:52,882 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:42:53,314 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:42:53,315 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:42:53,320 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:42:53,328 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:42:53,357 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:42:53,534 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:42:53,535 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:42:53,550 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:42:55,109 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:43:00,466 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:01,958 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:02,675 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:02,785 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:03,397 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:03,531 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:43:04,622 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:43:05,590 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:43:05,591 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:43:05,591 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:43:05,591 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:05,591 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer5_mpk.json, llima-compile
2026-03-07 05:43:05,777 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer5.sima
2026-03-07 05:43:05,851 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:43:05,851 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer5"
2026-03-07 05:43:05,853 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:43:05,853 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:43:05,879 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:43:05,880 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:43:05,880 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:43:05,880 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:43:05,921 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:43:05,926 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:43:05,937 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:43:06,062 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:06,183 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:43:06,505 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:43:06,507 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:43:06,515 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:43:13,633 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:13,643 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:43:13,643 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:43:13,669 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:43:14,745 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:15,819 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:16,350 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:16,397 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:16,724 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:16,944 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:17,584 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:17,731 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.945
2026-03-07 05:43:17,731 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:43:17,746 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:43:17,764 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:17,794 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:17,896 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:43:17,897 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:43:18,171 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:18,764 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:18,861 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:19,050 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer4_conv_stage1_mla.elf, llima-compile
2026-03-07 05:43:19,771 - mlc.test_util.test_context - INFO - Compression done in 16s. Compression ratio: 0.961
2026-03-07 05:43:19,771 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:43:19,800 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv.sima
2026-03-07 05:43:19,905 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:43:19,905 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv"
2026-03-07 05:43:19,910 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:43:19,910 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:43:19,961 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:43:19,961 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:43:19,961 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:43:19,961 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:43:19,990 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:43:20,355 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:43:20,356 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:43:23,896 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:43:23,901 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:43:24,169 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:43:24,183 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:43:25,389 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.98
2026-03-07 05:43:25,389 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:43:25,596 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:43:25,711 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:43:25,711 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:43:26,738 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:43:27,180 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:43:27,250 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:43:27,622 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:43:27,623 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:43:27,639 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:28,850 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer5_stage1_mla.elf, llima-compile
2026-03-07 05:43:29,329 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv.sima
2026-03-07 05:43:29,416 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:43:29,416 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv"
2026-03-07 05:43:29,419 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:43:29,419 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:43:29,455 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:43:29,456 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:43:29,456 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:43:29,456 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:43:29,531 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:43:29,559 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:43:29,580 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:43:29,803 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:43:29,804 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:43:29,817 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:43:34,285 - mlc.test_util.test_context - INFO - Compression done in 15s. Compression ratio: 0.963
2026-03-07 05:43:34,285 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:43:34,469 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:43:34,778 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:43:34,778 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:43:37,490 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:37,534 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:38,285 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer3_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:43:38,988 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv.sima
2026-03-07 05:43:39,065 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:43:39,065 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv"
2026-03-07 05:43:39,070 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:43:39,070 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:43:39,117 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:43:39,118 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:43:39,118 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:43:39,118 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:43:41,297 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:42,441 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:42,813 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:42,974 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:43:43,406 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:43,448 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:45,438 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:43:45,870 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:43:45,937 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:43:46,038 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:43:46,038 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:43:46,287 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:43:46,288 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:43:46,303 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:43:50,497 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:43:50,498 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer5_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer5_mpk.json, llima-compile
2026-03-07 05:43:50,996 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv.sima
2026-03-07 05:43:51,082 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:43:51,082 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv"
2026-03-07 05:43:51,085 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:43:51,085 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:43:51,121 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:43:51,122 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:43:51,122 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:43:51,122 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:43:51,197 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:43:51,224 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:43:51,246 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:43:51,522 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:43:51,523 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:43:51,541 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:43:52,356 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.976
2026-03-07 05:43:52,356 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:43:52,598 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:43:52,731 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:43:52,731 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:43:56,278 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:43:57,804 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:43:58,545 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:43:58,654 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:43:59,797 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:43:59,843 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:44:01,257 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:44:01,408 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:44:03,711 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:44:05,621 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:44:05,850 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:44:05,893 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:44:07,434 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:44:07,448 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:44:14,588 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:44:14,913 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.982
2026-03-07 05:44:14,913 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:44:15,159 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:44:15,299 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:44:15,299 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:44:15,799 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:44:15,811 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 10s.
2026-03-07 05:44:15,949 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.962
2026-03-07 05:44:15,949 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:44:16,082 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:44:16,166 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:44:16,386 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:44:16,400 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:44:16,533 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:44:16,533 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:44:16,818 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:44:16,929 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:20,871 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer4_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:44:21,597 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv.sima
2026-03-07 05:44:21,686 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:44:21,686 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv"
2026-03-07 05:44:21,691 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:44:21,692 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:44:21,738 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:44:21,739 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:44:21,739 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:44:21,739 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:22,371 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer6_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:44:22,887 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv.sima
2026-03-07 05:44:22,971 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:44:22,971 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv"
2026-03-07 05:44:22,974 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:44:22,974 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:44:23,010 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:44:23,010 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:44:23,010 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:44:23,010 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:44:23,086 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:44:23,114 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:44:23,136 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:44:23,381 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:44:23,382 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:44:23,395 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:26,826 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer5_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_post_layer5_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer5_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:44:27,233 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9.sima
2026-03-07 05:44:27,256 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:44:27,256 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9"
2026-03-07 05:44:27,259 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:44:27,259 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:44:27,264 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:44:27,265 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:44:27,265 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:44:27,265 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:44:28,045 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:44:28,471 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:44:28,537 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:44:28,857 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:44:28,858 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:44:28,874 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:44:31,627 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:44:31,677 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:44:33,954 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:44:34,287 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.967
2026-03-07 05:44:34,287 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:44:34,390 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:44:34,426 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:44:34,506 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:44:34,597 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:44:34,598 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:44:34,613 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:44:34,859 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:44:34,859 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:44:35,414 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:44:36,546 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:44:36,774 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:44:36,815 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:44:39,010 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:44:39,024 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:44:43,903 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:44:44,032 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:44:44,295 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:44:44,445 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:44:44,505 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer7_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:44:45,047 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer9.sima
2026-03-07 05:44:45,132 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:44:45,133 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer9"
2026-03-07 05:44:45,138 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:44:45,138 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:44:45,159 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:44:45,160 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:44:45,160 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:44:45,160 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:44:45,835 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.977
2026-03-07 05:44:45,835 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:44:46,076 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:44:46,212 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:44:46,212 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:44:50,894 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:44:50,966 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:44:51,018 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:44:51,445 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:44:51,446 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:44:51,458 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:44:54,803 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:44:55,855 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:44:56,363 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:44:56,399 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:44:56,408 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:44:57,731 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.95
2026-03-07 05:44:57,731 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:44:57,746 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:44:57,898 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:44:57,898 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:44:57,900 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:44:58,621 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:44:58,733 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:03,940 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:04,061 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:04,160 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:04,174 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:45:04,719 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:04,723 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:09,654 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:09,655 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer9_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:45:09,794 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:09,794 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:45:10,061 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9.sima
2026-03-07 05:45:10,080 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:45:10,080 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9"
2026-03-07 05:45:10,084 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:45:10,084 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:45:10,089 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:45:10,089 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:45:10,089 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:45:10,089 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:45:10,262 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:45:10,283 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:45:10,317 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:45:10,342 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:45:10,343 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:45:10,352 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:45:12,457 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:12,491 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:14,074 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:45:15,023 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:15,238 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer8_conv_stage1_mla.elf, llima-compile
2026-03-07 05:45:15,271 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:45:15,378 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:45:15,421 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer9.sima
2026-03-07 05:45:15,479 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:45:15,480 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer9"
2026-03-07 05:45:15,481 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:45:15,482 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:45:15,495 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:45:15,507 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:15,508 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:45:15,509 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:45:15,509 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:45:15,509 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:45:15,551 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:45:15,555 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:45:15,567 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:45:15,758 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:45:15,759 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:45:15,768 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:45:15,863 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:45:15,956 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:16,101 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.959
2026-03-07 05:45:16,101 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:45:16,226 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.974
2026-03-07 05:45:16,226 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:45:16,254 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:45:16,293 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:45:16,294 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:45:16,316 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:45:16,723 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:45:16,723 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:45:16,898 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:16,899 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:16,899 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer6_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:45:17,618 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv.sima
2026-03-07 05:45:17,697 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:45:17,698 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv"
2026-03-07 05:45:17,702 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:45:17,703 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:45:17,751 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:45:17,751 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:45:17,751 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:45:17,751 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:45:18,735 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:22,635 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:22,670 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:22,879 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:22,893 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:45:23,996 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:45:24,426 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:45:24,494 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:45:24,837 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:45:24,838 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:45:24,855 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:45:25,655 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:45:26,518 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:45:26,697 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:45:26,729 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:29,205 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:30,167 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer9_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:45:30,430 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv.sima
2026-03-07 05:45:30,508 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:45:30,508 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv"
2026-03-07 05:45:30,510 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:45:30,511 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:45:30,527 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:45:30,528 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:45:30,528 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:45:30,528 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:45:30,603 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:45:30,631 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:45:30,652 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:45:30,688 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.96
2026-03-07 05:45:30,688 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:45:30,874 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:45:30,880 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:45:30,881 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:45:30,894 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:45:31,332 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:45:31,332 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:45:34,953 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.978
2026-03-07 05:45:34,953 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:45:35,157 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:45:35,263 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:45:35,263 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:35,276 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer7_conv_stage1_mla.elf, llima-compile
2026-03-07 05:45:36,013 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv.sima
2026-03-07 05:45:36,098 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:45:36,098 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv"
2026-03-07 05:45:36,103 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:45:36,103 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:45:36,140 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:45:36,141 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:45:36,141 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:45:36,141 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:45:38,982 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:39,025 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:40,204 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:40,350 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:42,517 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:45:42,728 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:45:42,944 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:45:43,010 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:45:43,335 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:45:43,336 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:45:43,351 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:45:43,863 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:45:44,096 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:45:44,136 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:52,131 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:45:53,600 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:45:53,794 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.981
2026-03-07 05:45:53,794 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:45:54,036 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:45:54,174 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:45:54,174 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:45:54,311 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:45:54,417 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:45:55,101 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:45:55,114 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:45:57,936 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:45:58,084 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:45:59,429 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer9_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer9_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer9_mpk.json, llima-compile
2026-03-07 05:45:59,927 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv.sima
2026-03-07 05:46:00,001 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:46:00,001 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv"
2026-03-07 05:46:00,004 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:46:00,004 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:46:00,030 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:46:00,031 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:46:00,031 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:46:00,031 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:46:00,106 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:46:00,134 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:46:00,155 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:46:00,384 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:46:00,386 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:46:00,400 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:46:05,199 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:46:05,215 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:46:08,499 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:46:08,543 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:46:10,790 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:46:11,341 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:46:11,341 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 10s.
2026-03-07 05:46:11,634 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.965
2026-03-07 05:46:11,634 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:46:11,847 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:46:12,207 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:46:12,207 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:46:12,288 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:46:12,292 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:46:13,006 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:46:13,114 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:46:13,463 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:46:13,691 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:46:13,733 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:17,703 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer8_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:46:17,833 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:46:17,847 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:46:18,417 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv.sima
2026-03-07 05:46:18,498 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:46:18,498 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv"
2026-03-07 05:46:18,503 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:46:18,503 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:46:18,539 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:46:18,539 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:46:18,539 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:46:18,540 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:21,666 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:46:21,667 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:46:21,667 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:46:21,667 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:46:21,667 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:21,667 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer9_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer9_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer9_stage1_mla.elf, llima-compile
2026-03-07 05:46:22,150 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv.sima
2026-03-07 05:46:22,254 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:46:22,254 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv"
2026-03-07 05:46:22,257 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:46:22,257 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:46:22,291 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:46:22,292 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:46:22,292 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:46:22,292 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:46:22,366 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:46:22,394 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:46:22,414 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:46:22,639 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:46:22,639 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:46:22,741 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.979
2026-03-07 05:46:22,742 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:46:22,986 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:46:23,071 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:46:23,124 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:46:23,124 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:23,202 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer10_conv_stage1_mla.elf, llima-compile
2026-03-07 05:46:23,619 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13.sima
2026-03-07 05:46:23,660 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:46:23,660 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13"
2026-03-07 05:46:23,663 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:46:23,663 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:46:23,668 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:46:23,669 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:46:23,669 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:46:23,669 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:46:24,893 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:46:25,324 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:46:25,390 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:46:25,725 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:46:25,725 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:46:25,741 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:46:30,444 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.961
2026-03-07 05:46:30,444 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:46:30,539 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:46:30,662 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:46:30,980 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:46:31,017 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:46:31,026 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:46:31,026 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:46:31,158 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:46:31,192 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:46:31,193 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:46:31,202 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:46:31,208 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:46:35,038 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:46:36,177 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:46:36,404 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:46:36,446 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:46:40,258 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:46:40,406 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:46:40,858 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:46:40,992 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:46:45,492 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.973
2026-03-07 05:46:45,492 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:46:45,732 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:46:45,867 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:46:45,868 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:46:46,785 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:46:46,786 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:46:51,976 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:46:52,167 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:46:52,168 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:46:52,168 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer11_conv_mpk.json, llima-compile
2026-03-07 05:46:52,769 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer13.sima
2026-03-07 05:46:52,839 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:46:52,839 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer13"
2026-03-07 05:46:52,843 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:46:52,843 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:46:52,870 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:46:52,870 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:46:52,870 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:46:52,870 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:46:53,026 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:46:53,104 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:46:53,540 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:46:53,587 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:46:54,607 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:46:54,919 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.948
2026-03-07 05:46:54,919 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:46:54,934 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:46:55,084 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:46:55,084 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:46:55,318 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:46:55,427 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:46:59,110 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:46:59,181 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:46:59,234 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:46:59,659 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:46:59,660 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:46:59,673 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:00,679 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:00,695 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:47:01,060 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:01,063 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:06,071 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer13_stage1_mla.elf, llima-compile
2026-03-07 05:47:06,475 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13.sima
2026-03-07 05:47:06,498 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:06,498 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13"
2026-03-07 05:47:06,502 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:06,502 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:06,507 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:06,507 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:06,507 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:06,507 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:06,681 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:06,702 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:06,735 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:06,759 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:06,760 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:06,769 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:09,339 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:09,370 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:09,698 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:09,698 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:47:11,432 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:47:11,781 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:11,784 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:47:11,899 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:47:11,909 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:11,911 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:47:12,628 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.969
2026-03-07 05:47:12,628 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:47:12,656 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:47:12,695 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:47:12,695 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:47:12,771 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.952
2026-03-07 05:47:12,771 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:47:12,985 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:13,114 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer10_conv_stage1_mla.elf, llima-compile
2026-03-07 05:47:13,271 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer13.sima
2026-03-07 05:47:13,332 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:13,332 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer13"
2026-03-07 05:47:13,334 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:13,334 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:13,344 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:47:13,344 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:47:13,347 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:13,348 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:13,348 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:13,348 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:13,388 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:13,392 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:13,404 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:13,588 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:13,590 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:13,598 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:15,036 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer12_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:47:15,054 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:16,174 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv.sima
2026-03-07 05:47:16,264 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:16,264 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv"
2026-03-07 05:47:16,269 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:16,269 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:16,320 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:16,321 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:16,321 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:16,321 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:18,279 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:18,293 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:47:20,082 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:20,117 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:22,558 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:47:22,629 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:23,053 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:23,107 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:47:23,120 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:23,494 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:23,495 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:23,511 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:23,761 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:47:23,964 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:47:24,362 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:47:24,457 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:47:24,851 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:47:24,884 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:47:27,014 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:27,969 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer13_stage1_mla.elf, llima-compile
2026-03-07 05:47:28,267 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv.sima
2026-03-07 05:47:28,337 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:28,337 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv"
2026-03-07 05:47:28,339 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:28,340 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:28,366 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:28,366 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:28,366 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:28,366 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:28,441 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:28,469 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:28,489 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:28,742 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:28,743 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:28,757 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:30,616 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:30,617 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer11_conv_stage1_mla.elf, llima-compile
2026-03-07 05:47:31,367 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv.sima
2026-03-07 05:47:31,447 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:31,447 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv"
2026-03-07 05:47:31,452 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:31,452 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:31,499 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:31,499 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:31,499 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:31,499 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:32,440 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.978
2026-03-07 05:47:32,440 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:47:32,641 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:47:32,748 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:47:32,749 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:47:37,444 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:37,487 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:38,219 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:38,646 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:38,712 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:38,787 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:38,938 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:39,042 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:39,043 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:39,059 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:47:39,259 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.959
2026-03-07 05:47:39,260 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:47:39,443 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:47:39,744 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:47:39,744 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:47:41,208 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:47:42,334 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:47:42,561 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:47:42,601 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:47:51,490 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.98
2026-03-07 05:47:51,490 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:47:51,732 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:47:51,869 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:47:51,869 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:47:52,257 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:47:52,794 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:47:52,808 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:47:53,736 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:47:53,749 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:47:53,886 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:47:54,479 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:47:54,589 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:47:57,159 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:47:57,160 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer13_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_post_layer13_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer13_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:47:57,651 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv.sima
2026-03-07 05:47:57,737 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:47:57,737 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv"
2026-03-07 05:47:57,740 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:47:57,740 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:47:57,775 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:47:57,776 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:47:57,776 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:47:57,776 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:47:57,851 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:47:57,881 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:47:57,905 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:47:58,145 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:47:58,145 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:47:58,159 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:48:01,531 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:48:01,546 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:48:06,361 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:48:06,405 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:48:07,102 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:48:08,583 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:48:09,289 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:48:09,395 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:48:10,092 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:48:11,216 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:48:11,437 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:48:11,476 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:48:11,874 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.963
2026-03-07 05:48:11,874 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:48:12,093 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:48:12,467 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:48:12,468 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:13,892 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer12_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:48:14,588 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv.sima
2026-03-07 05:48:14,669 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:48:14,670 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv"
2026-03-07 05:48:14,674 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:48:14,674 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:48:14,722 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:48:14,723 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:48:14,723 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:48:14,723 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:48:15,705 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:48:15,705 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:48:19,433 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:48:19,445 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 11s.
2026-03-07 05:48:20,378 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.978
2026-03-07 05:48:20,378 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:48:20,616 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:48:20,748 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:48:20,749 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:21,032 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer14_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:48:21,330 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv.sima
2026-03-07 05:48:21,398 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:48:21,399 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv"
2026-03-07 05:48:21,401 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:48:21,401 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:48:21,421 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:48:21,436 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:48:21,436 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:48:21,437 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:48:21,437 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:48:21,511 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:48:21,539 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:48:21,560 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:48:21,792 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:48:21,793 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:48:21,806 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:48:21,847 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:48:21,915 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:48:22,251 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:48:22,252 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:48:22,267 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:48:26,762 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.961
2026-03-07 05:48:26,762 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:48:26,979 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:48:27,332 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:48:27,332 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:48:30,276 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:48:30,276 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:30,277 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer13_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer13_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_post_layer13_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:48:30,409 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:48:30,453 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:48:30,697 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17.sima
2026-03-07 05:48:30,717 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:48:30,717 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17"
2026-03-07 05:48:30,720 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:48:30,720 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:48:30,725 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:48:30,726 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:48:30,726 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:48:30,726 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:48:34,254 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:48:35,390 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:48:35,618 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:48:35,660 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:48:37,264 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:48:37,383 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:48:37,416 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:48:37,815 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:48:37,851 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:48:38,021 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:48:38,022 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:48:38,037 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:48:44,475 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:48:44,475 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:48:44,671 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.974
2026-03-07 05:48:44,671 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:48:44,911 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:48:45,047 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:48:45,047 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:48:47,897 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:48:48,031 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:48:49,824 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:48:49,825 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer15_conv_stage1_mla.elf, llima-compile
2026-03-07 05:48:50,223 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer17.sima
2026-03-07 05:48:50,295 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:48:50,295 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer17"
2026-03-07 05:48:50,299 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:48:50,299 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:48:50,320 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:48:50,320 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:48:50,320 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:48:50,320 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:48:50,595 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:48:52,119 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:48:52,835 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:48:52,943 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:48:56,470 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:48:56,541 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:48:56,593 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:48:57,011 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:48:57,012 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:48:57,025 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:48:58,095 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:48:59,135 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:00,482 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:00,497 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:49:00,590 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:00,635 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:01,964 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.952
2026-03-07 05:49:01,964 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:01,979 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:02,127 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:02,127 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:08,135 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:08,619 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:08,619 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:49:09,672 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:49:09,717 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:09,838 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:10,279 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.953
2026-03-07 05:49:10,279 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:10,495 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:10,853 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:10,853 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:12,881 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer14_conv_stage1_mla.elf, llima-compile
2026-03-07 05:49:13,283 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17.sima
2026-03-07 05:49:13,302 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:13,302 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17"
2026-03-07 05:49:13,305 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:13,305 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:13,310 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:13,311 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:13,311 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:13,311 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:13,483 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:13,503 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:13,536 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:13,562 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:13,562 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:13,571 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:13,941 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer16_conv_stage1_mla.elf, llima-compile
2026-03-07 05:49:14,124 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer17.sima
2026-03-07 05:49:14,181 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:14,182 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer17"
2026-03-07 05:49:14,183 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:14,183 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:14,200 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:14,201 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:14,201 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:14,201 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:14,241 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:14,245 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:14,257 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:14,262 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer17_mpk.json, llima-compile
2026-03-07 05:49:14,442 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:14,443 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:14,451 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:49:14,805 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv.sima
2026-03-07 05:49:14,884 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:14,884 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv"
2026-03-07 05:49:14,889 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:14,889 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:14,927 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:14,927 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:14,927 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:14,928 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:15,057 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:15,075 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:49:16,130 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:16,162 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:18,282 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:49:18,640 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:18,757 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:18,769 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:19,483 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.972
2026-03-07 05:49:19,483 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:19,512 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:19,552 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:19,553 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:19,739 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:49:20,943 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:21,337 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:21,425 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:21,460 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:21,532 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:21,624 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:21,764 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:21,830 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:21,987 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:22,197 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:22,198 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:22,214 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:49:24,447 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:49:25,304 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:25,484 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:25,517 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:27,565 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:27,566 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer15_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:49:27,865 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv.sima
2026-03-07 05:49:27,939 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:27,939 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv"
2026-03-07 05:49:27,942 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:27,942 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:27,972 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:27,973 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:27,973 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:27,973 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:28,047 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:28,074 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:28,095 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:28,322 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:28,323 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:28,336 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:49:31,252 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:32,205 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer17_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:49:32,915 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv.sima
2026-03-07 05:49:33,023 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:33,023 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv"
2026-03-07 05:49:33,027 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:33,028 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:33,053 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:33,053 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:33,054 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:33,054 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:33,785 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.977
2026-03-07 05:49:33,785 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:33,989 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:34,097 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:34,097 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:36,024 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:36,068 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:36,491 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.957
2026-03-07 05:49:36,491 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:36,677 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:36,983 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:36,984 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:37,043 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:37,197 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:39,703 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:39,868 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:49:40,132 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:40,198 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:40,565 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:40,566 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:40,581 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:49:41,768 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:42,002 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:42,043 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:49,221 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:49:50,754 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:49:50,969 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.978
2026-03-07 05:49:50,970 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:49:51,211 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:49:51,344 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:49:51,344 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:49:52,762 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:49:52,869 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:49:54,019 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:54,030 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:49:55,597 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:49:55,749 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:49:58,429 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:49:58,430 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer17_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer17_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer17_mpk.json, llima-compile
2026-03-07 05:49:58,516 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:49:58,530 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:49:58,901 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv.sima
2026-03-07 05:49:58,970 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:49:58,970 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv"
2026-03-07 05:49:58,973 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:49:58,973 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:49:59,001 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:49:59,002 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:49:59,002 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:49:59,002 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:49:59,076 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:49:59,105 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:49:59,126 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:49:59,342 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:49:59,342 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:49:59,356 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:50:07,679 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:50:07,722 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:50:07,879 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:50:10,186 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.959
2026-03-07 05:50:10,186 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:50:10,402 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:50:10,740 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:50:10,771 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:50:10,771 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:50:10,929 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:10,930 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer16_conv_mpk.json, llima-compile
2026-03-07 05:50:11,454 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:50:11,457 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:50:11,566 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:50:11,632 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv.sima
2026-03-07 05:50:11,714 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:50:11,714 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv"
2026-03-07 05:50:11,719 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:50:11,719 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:50:11,770 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:50:11,770 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:50:11,770 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:50:11,770 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:50:12,583 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:50:12,809 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:50:12,850 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:50:14,977 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:50:14,992 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:50:17,515 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:50:17,515 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 11s.
2026-03-07 05:50:18,074 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:50:18,506 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:50:18,573 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:50:18,906 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:50:18,908 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:50:18,924 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:20,324 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer18_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:50:20,590 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv.sima
2026-03-07 05:50:20,667 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:50:20,667 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv"
2026-03-07 05:50:20,670 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:50:20,670 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:50:20,687 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:50:20,688 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:50:20,688 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:50:20,688 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:50:20,763 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:50:20,790 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:50:20,811 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:50:21,032 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:50:21,033 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:50:21,046 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:50:21,817 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.977
2026-03-07 05:50:21,817 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:50:22,060 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:50:22,194 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:50:22,195 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:28,160 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer17_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer17_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_post_layer17_mpk.json, llima-compile
2026-03-07 05:50:28,857 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.958
2026-03-07 05:50:28,857 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:50:29,028 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21.sima
2026-03-07 05:50:29,051 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:50:29,051 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21"
2026-03-07 05:50:29,056 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:50:29,056 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:50:29,061 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:50:29,061 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:50:29,061 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:50:29,061 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:50:29,073 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:50:29,136 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:50:29,180 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:50:29,441 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:50:29,441 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:50:32,893 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:50:33,882 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:50:34,017 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:50:34,034 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:50:34,243 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:50:34,284 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:50:35,743 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:50:36,179 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:50:36,216 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:50:36,388 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:50:36,389 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:50:36,404 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:50:43,202 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.973
2026-03-07 05:50:43,202 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:50:43,444 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:50:43,577 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:50:43,577 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:50:45,720 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:50:45,851 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:50:45,938 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:50:45,938 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:50:47,029 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:50:48,538 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:50:49,265 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:50:49,377 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:50:51,350 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer19_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:50:51,770 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer21.sima
2026-03-07 05:50:51,838 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:50:51,838 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer21"
2026-03-07 05:50:51,842 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:50:51,842 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:50:51,865 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:50:51,866 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:50:51,866 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:50:51,866 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:50:56,511 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:50:57,405 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:50:57,420 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:50:57,550 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:50:58,036 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:50:58,055 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:50:58,100 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:50:58,107 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:50:58,158 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:50:58,572 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:50:58,573 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:50:58,585 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:50:59,434 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.935
2026-03-07 05:50:59,435 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:50:59,449 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:50:59,598 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:50:59,598 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:05,633 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:05,639 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:51:06,815 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.952
2026-03-07 05:51:06,815 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:51:07,036 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:51:07,400 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:51:07,400 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:07,888 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:07,889 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:09,692 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer18_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:51:10,101 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21.sima
2026-03-07 05:51:10,120 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:10,121 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21"
2026-03-07 05:51:10,124 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:10,124 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:10,129 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:10,130 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:10,130 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:10,130 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:10,140 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer21_mpk.json, llima-compile
2026-03-07 05:51:10,304 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:10,325 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:10,359 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:10,383 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:10,384 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:10,394 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:51:10,491 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer21.sima
2026-03-07 05:51:10,547 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:10,547 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer21"
2026-03-07 05:51:10,549 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:10,549 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:10,567 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:10,568 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:10,568 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:10,568 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:10,609 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:10,613 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:10,624 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:10,665 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:10,787 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:10,810 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:10,811 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:10,819 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:51:12,492 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:12,524 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:13,190 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer20_conv_stage1_mla.elf, llima-compile
2026-03-07 05:51:14,348 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv.sima
2026-03-07 05:51:14,424 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:14,424 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv"
2026-03-07 05:51:14,429 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:14,429 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:14,455 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:14,455 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:14,455 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:14,455 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:15,142 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:51:15,498 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:51:15,613 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:51:15,624 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:51:16,383 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.959
2026-03-07 05:51:16,383 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:51:16,414 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:51:16,453 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:51:16,454 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:16,825 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:16,839 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:51:17,638 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:17,673 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:18,830 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:20,663 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:20,685 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:51:21,089 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:21,155 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:21,499 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:21,499 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:21,515 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:51:21,543 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:51:21,546 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:51:21,728 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:51:21,761 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:51:22,734 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:51:23,320 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:51:23,413 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:51:29,328 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:29,329 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer19_conv_mpk.json, llima-compile
2026-03-07 05:51:29,372 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.974
2026-03-07 05:51:29,372 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:51:29,577 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:51:29,580 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:51:29,686 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:51:29,687 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:29,821 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv.sima
2026-03-07 05:51:29,900 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:29,900 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv"
2026-03-07 05:51:29,903 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:29,903 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:29,938 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:29,939 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:29,939 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:29,939 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:30,014 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:30,042 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:30,063 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:30,289 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:30,289 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:30,303 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:30,542 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer21_mpk.json, llima-compile
2026-03-07 05:51:31,248 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv.sima
2026-03-07 05:51:31,328 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:31,329 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv"
2026-03-07 05:51:31,333 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:31,333 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:31,359 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:31,360 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:31,360 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:31,360 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:36,563 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:36,712 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:37,608 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:38,036 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:38,074 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:38,103 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:38,118 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:38,179 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.953
2026-03-07 05:51:38,179 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:51:38,360 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:51:38,508 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:38,508 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:38,672 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:51:38,673 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:38,878 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:51:41,930 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:51:43,070 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:51:43,298 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:51:43,339 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:51:49,732 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:51:49,847 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:49,847 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:51:51,204 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:51:51,924 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:51:52,032 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:51:53,173 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.978
2026-03-07 05:51:53,173 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:51:53,413 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:51:53,548 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:51:53,548 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:51:53,663 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:51:53,813 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:51:54,248 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer21_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_post_layer21_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer21_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:51:54,740 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv.sima
2026-03-07 05:51:54,817 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:51:54,818 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv"
2026-03-07 05:51:54,820 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:51:54,821 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:51:54,855 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:51:54,859 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:51:54,875 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:51:55,272 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:51:55,272 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:51:55,272 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:51:55,346 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:51:55,374 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:51:55,395 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:51:55,591 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:51:55,592 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:51:55,605 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:03,812 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:03,856 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:06,918 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:07,097 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer20_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:52:07,500 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24.sima
2026-03-07 05:52:07,520 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:07,520 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24"
2026-03-07 05:52:07,524 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:07,524 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:07,529 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:07,530 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:07,530 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:07,530 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:07,597 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:08,396 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:08,736 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:08,967 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:09,007 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:52:09,107 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:09,214 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:52:09,400 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.959
2026-03-07 05:52:09,400 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:52:09,615 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:52:09,973 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:52:09,973 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:52:14,332 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:14,778 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:14,814 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:14,985 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:14,986 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:15,000 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:17,244 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:17,258 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:52:18,121 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.976
2026-03-07 05:52:18,121 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:52:18,856 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:18,869 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 11s.
2026-03-07 05:52:19,167 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:52:19,302 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:52:19,302 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:22,589 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer22_conv_mpk.json, llima-compile
2026-03-07 05:52:23,167 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer24.sima
2026-03-07 05:52:23,234 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:23,234 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer24"
2026-03-07 05:52:23,238 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:23,238 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:23,259 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:23,259 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:23,259 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:23,259 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:24,338 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:24,469 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:26,565 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.957
2026-03-07 05:52:26,565 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:52:26,779 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:52:27,131 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:52:27,131 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:52:29,020 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:29,093 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:29,118 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer21_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer21_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer21_stage1_mla.elf, llima-compile
2026-03-07 05:52:29,144 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:29,518 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24.sima
2026-03-07 05:52:29,538 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:29,538 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24"
2026-03-07 05:52:29,541 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:29,541 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:29,546 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:29,547 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:29,547 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:29,547 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:29,594 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:29,595 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:29,608 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:29,718 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:29,739 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:29,772 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:29,795 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:29,796 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:29,806 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:32,325 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:32,356 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:34,399 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:34,750 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:34,862 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:34,873 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:52:35,388 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:35,583 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.965
2026-03-07 05:52:35,583 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:52:35,614 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:52:35,653 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:52:35,653 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:52:36,450 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:36,976 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:37,023 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:52:37,995 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:38,365 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.937
2026-03-07 05:52:38,365 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:52:38,379 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:52:38,530 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:52:38,530 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:52:42,301 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:42,426 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:43,219 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:43,234 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:52:44,477 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:44,480 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:52:46,203 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:47,193 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer24_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:52:47,516 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer24.sima
2026-03-07 05:52:47,574 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:47,574 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer24"
2026-03-07 05:52:47,576 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:47,576 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:47,602 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:47,602 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:47,602 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:47,602 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:47,643 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:47,647 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:47,659 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:47,842 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:47,843 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:47,851 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:48,564 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:48,565 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer23_conv_mpk.json, llima-compile
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:52:49,054 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer24_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:52:49,289 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv.sima
2026-03-07 05:52:49,350 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv.sima
2026-03-07 05:52:49,366 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:49,366 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv"
2026-03-07 05:52:49,371 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:49,371 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:49,404 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:49,405 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:49,405 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:49,405 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:49,419 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:52:49,419 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv"
2026-03-07 05:52:49,421 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:52:49,422 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:52:49,447 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:52:49,447 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:52:49,447 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:52:49,447 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:52:49,522 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:49,549 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:49,570 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:49,780 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:49,781 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:49,794 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:52,482 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:53,686 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:54,278 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:54,374 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:52:54,621 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:54,655 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:55,702 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:52:56,143 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:52:56,211 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:52:56,537 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:52:56,538 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:52:56,554 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:52:57,656 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:52:57,949 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:52:57,993 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:52:58,517 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:52:58,521 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:52:58,534 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:52:58,702 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:52:58,733 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:53:01,754 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:53:02,897 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:53:03,129 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:53:03,173 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:53:06,308 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.98
2026-03-07 05:53:06,308 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:53:06,514 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:53:06,622 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:53:06,623 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:53:09,106 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.961
2026-03-07 05:53:09,106 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:53:09,290 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:53:09,607 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:53:09,607 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:10,862 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer22_conv_stage1_mla.elf, llima-compile
2026-03-07 05:53:11,556 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:53:11,601 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv.sima
2026-03-07 05:53:11,683 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:53:11,683 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv"
2026-03-07 05:53:11,688 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:53:11,688 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:53:11,709 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:53:11,738 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:53:11,738 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:53:11,738 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:53:11,738 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:53:12,362 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.982
2026-03-07 05:53:12,362 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:53:12,607 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:53:12,745 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:53:12,745 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:53:13,943 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:53:13,957 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:53:18,061 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:53:18,490 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:53:18,558 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:53:18,884 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:53:18,884 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:53:18,899 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:53:24,463 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:25,980 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer23_conv_stage1_mla.elf, llima-compile
2026-03-07 05:53:25,993 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:53:26,276 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv.sima
2026-03-07 05:53:26,354 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:53:26,354 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv"
2026-03-07 05:53:26,357 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:53:26,357 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:53:26,388 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:53:26,388 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:53:26,388 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:53:26,388 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:53:26,462 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:53:26,490 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:53:26,511 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:53:26,724 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:53:26,728 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:53:26,728 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:53:26,742 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:53:26,833 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:53:26,899 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:53:26,899 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:31,337 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer24_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer24_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer24_mpk.json, llima-compile
2026-03-07 05:53:32,148 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27.sima
2026-03-07 05:53:32,170 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:53:32,170 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27"
2026-03-07 05:53:32,174 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:53:32,174 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:53:32,179 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:53:32,179 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:53:32,179 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:53:32,179 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:53:33,632 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:53:33,782 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:53:34,908 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:53:34,952 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:53:36,598 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:53:36,599 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:53:38,691 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:53:38,847 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:53:39,279 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:53:39,316 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:53:39,484 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:53:39,485 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:53:39,499 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:53:39,826 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:53:40,054 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:53:40,096 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:41,872 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer25_conv_stage1_mla.elf, llima-compile
2026-03-07 05:53:42,864 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_post_layer27.sima
2026-03-07 05:53:42,930 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:53:42,930 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_post_layer27"
2026-03-07 05:53:42,934 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:53:42,934 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:53:42,959 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:53:42,960 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:53:42,960 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:53:42,960 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:53:44,153 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.965
2026-03-07 05:53:44,153 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:53:44,367 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:53:44,728 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:53:44,728 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:53:46,606 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:53:48,085 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:53:48,633 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:53:48,705 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:53:48,756 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:53:48,803 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:53:48,805 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:53:48,913 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:53:48,935 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:53:49,052 - mlc.test_util.test_context - INFO - Compression done in 8s. Compression ratio: 0.982
2026-03-07 05:53:49,052 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:53:49,212 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:53:49,213 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:53:49,225 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:53:49,297 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:53:49,432 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:53:49,432 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:53:49,452 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:53:49,464 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 10s.
2026-03-07 05:53:59,731 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:53:59,783 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer24_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_post_layer24_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer24_stage1_mla.elf, llima-compile
2026-03-07 05:54:00,186 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27.sima
2026-03-07 05:54:00,206 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:00,206 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27"
2026-03-07 05:54:00,210 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:00,210 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:00,215 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:00,215 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:00,215 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:00,215 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:00,388 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:00,408 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:00,442 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:00,466 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:00,467 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:00,476 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:00,776 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:01,289 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:01,335 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:01,734 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:01,855 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:02,678 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.924
2026-03-07 05:54:02,678 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:02,692 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:02,843 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:02,843 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:02,972 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:03,003 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:05,576 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:05,931 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:06,042 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:06,053 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:06,147 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.966
2026-03-07 05:54:06,147 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:06,363 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:06,720 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:06,720 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:06,772 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.957
2026-03-07 05:54:06,772 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:06,803 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:06,842 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:06,843 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:08,733 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:08,737 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:54:09,198 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:11,641 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:12,844 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:13,256 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_pre_layer27_stage1_mla.elf, llima-compile
2026-03-07 05:54:13,442 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:13,450 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:13,451 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:54:13,542 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:13,587 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer27.sima
2026-03-07 05:54:13,651 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:13,651 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer27"
2026-03-07 05:54:13,653 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:13,653 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:13,680 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:13,680 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:13,680 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:13,680 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:13,720 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:13,725 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:13,736 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:13,916 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:13,917 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:13,925 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:16,940 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - 	EV74: 7
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:17,900 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_pre_layer27_stage1_mla.elf, llima-compile
2026-03-07 05:54:18,437 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv.sima
2026-03-07 05:54:18,515 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:18,515 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv"
2026-03-07 05:54:18,520 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:18,520 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:18,554 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:18,555 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:18,555 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:18,555 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:54:18,769 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:18,770 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer26_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:54:19,255 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv.sima
2026-03-07 05:54:19,341 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:19,341 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv"
2026-03-07 05:54:19,343 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:19,344 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:19,364 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:19,365 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:19,365 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:19,365 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:19,441 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:19,469 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:19,490 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:19,707 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:19,708 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:19,721 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:20,696 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:20,730 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:23,718 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:24,578 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:24,755 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:24,787 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:25,277 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:25,707 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:25,773 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:26,144 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:26,145 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:26,160 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:27,827 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:27,871 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:28,354 - mlc.test_util.test_context - INFO - Compression done in 14s. Compression ratio: 0.964
2026-03-07 05:54:28,354 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:29,693 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:30,021 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:30,021 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:31,617 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:32,325 - mlc.test_util.test_context - INFO - Compression done in 7s. Compression ratio: 0.981
2026-03-07 05:54:32,325 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:32,452 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:32,466 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 15s.
2026-03-07 05:54:32,755 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:32,982 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:33,024 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:33,208 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:33,319 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:33,319 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:41,328 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:41,479 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:42,683 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.981
2026-03-07 05:54:42,683 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:42,922 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:43,057 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:43,057 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:54:44,722 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:44,723 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_layer25_conv_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:54:44,894 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv.sima
2026-03-07 05:54:44,914 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:44,914 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv"
2026-03-07 05:54:44,916 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:44,916 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:44,920 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:44,921 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:44,921 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:44,921 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:45,815 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:46,183 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:46,199 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:46,366 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:46,367 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:46,373 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:50,665 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:50,711 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:54:53,333 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:53,344 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:54:53,559 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:54,834 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:54:54,848 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:54:55,054 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:55,193 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:54:55,767 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:55,771 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:54:55,875 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:55,968 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:54:55,995 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:54:57,761 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer27_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_post_layer27_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer27_stage1_mla.elf, llima-compile
2026-03-07 05:54:57,933 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv.sima
2026-03-07 05:54:57,953 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:54:57,953 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv"
2026-03-07 05:54:57,955 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:54:57,955 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:54:57,960 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:54:57,960 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:54:57,960 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:54:57,960 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:54:58,013 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:54:58,038 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:54:58,053 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:54:58,089 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:54:58,090 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:54:58,100 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:54:59,523 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.952
2026-03-07 05:54:59,523 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:54:59,562 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:54:59,648 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:54:59,648 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:54:59,810 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:54:59,822 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:00,879 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:55:01,359 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:55:01,419 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:55:01,428 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:55:03,240 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.976
2026-03-07 05:55:03,240 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:55:03,286 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:55:03,321 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:55:03,322 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:55:06,917 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:06,930 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:07,013 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:07,014 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:07,014 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:55:07,014 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:07,014 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:07,014 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer26_conv_mpk.json, llima-compile
2026-03-07 05:55:07,287 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token0.sima
2026-03-07 05:55:07,301 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:07,301 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token0"
2026-03-07 05:55:07,305 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:07,305 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:07,306 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:07,306 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:07,306 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:07,306 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:08,230 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:09,484 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:09,493 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 5s.
2026-03-07 05:55:10,759 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:10,771 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 10s.
2026-03-07 05:55:11,311 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:55:11,524 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:11,536 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:11,744 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:11,744 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:11,749 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,226 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer28_conv_stage1_mla.elf, llima-compile
2026-03-07 05:55:12,502 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token128.sima
2026-03-07 05:55:12,514 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:12,514 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token128"
2026-03-07 05:55:12,517 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:12,517 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:12,518 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:12,518 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:12,518 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:12,518 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:12,850 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer29_conv_mpk.json, llima-compile
2026-03-07 05:55:13,121 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token256.sima
2026-03-07 05:55:13,131 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:13,131 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token256"
2026-03-07 05:55:13,134 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:13,134 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:13,135 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:13,136 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:13,136 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:13,136 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:13,202 - mlc.test_util.test_context - INFO - Compression done in 17s. Compression ratio: 0.963
2026-03-07 05:55:13,202 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:55:13,422 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:55:13,796 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:55:13,796 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:55:16,095 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:55:16,326 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:16,338 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:16,610 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:16,611 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:16,616 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:16,620 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 680
2026-03-07 05:55:16,621 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:55:18,697 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:55:19,905 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:19,906 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_layer29_conv_stage1_mla.elf, llima-compile
2026-03-07 05:55:20,174 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token384.sima
2026-03-07 05:55:20,184 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:20,184 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token384"
2026-03-07 05:55:20,187 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:20,188 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:20,189 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:20,189 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:20,189 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:20,189 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - 	EV74: 3
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:21,178 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_post_layer27_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_post_layer27_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_post_layer27_mpk.json, llima-compile
2026-03-07 05:55:21,446 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token512.sima
2026-03-07 05:55:21,456 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:21,456 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token512"
2026-03-07 05:55:21,460 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:21,460 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:21,461 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:21,461 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:21,461 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:21,461 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:23,048 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:55:23,245 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:23,262 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:23,618 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:23,619 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:23,626 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:24,217 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 933
2026-03-07 05:55:24,217 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:55:24,533 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 705
2026-03-07 05:55:24,533 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:55:25,009 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:55:25,078 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:30,376 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:55:30,506 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:30,522 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:30,757 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:55:30,941 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:30,942 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:30,949 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:31,354 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:55:31,605 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:55:31,623 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:55:31,639 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.555
2026-03-07 05:55:31,639 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:55:31,641 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:55:31,700 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:55:31,700 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:55:32,890 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:32,892 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:55:33,109 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:55:33,238 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:33,260 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:33,755 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:33,756 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:33,765 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:55:34,886 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:34,887 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:34,887 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token0_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token0_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token0_mpk.json, llima-compile
2026-03-07 05:55:35,159 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token640.sima
2026-03-07 05:55:35,169 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:35,169 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token640"
2026-03-07 05:55:35,171 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:35,171 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:35,172 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:35,172 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:35,172 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:35,172 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:35,494 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:55:35,601 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:38,198 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 694
2026-03-07 05:55:38,198 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:55:45,183 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:55:45,569 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:55:45,679 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:45,897 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:55:46,320 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:55:46,359 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:55:46,379 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.391
2026-03-07 05:55:46,379 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:55:46,385 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:55:46,497 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:55:46,497 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:55:48,930 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:48,932 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 05:55:49,239 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:55:49,532 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:55:49,561 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:55:50,128 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:55:50,129 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:55:50,141 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:55:53,032 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token128_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token128_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token128_mpk.json, llima-compile
2026-03-07 05:55:53,302 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token768.sima
2026-03-07 05:55:53,312 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:55:53,312 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token768"
2026-03-07 05:55:53,314 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:55:53,314 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:55:53,315 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:55:53,315 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:55:53,315 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:55:53,315 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:55:55,332 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:55:56,067 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:55:56,104 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:55:56,202 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:56,539 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:55:56,577 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:55:56,599 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.318
2026-03-07 05:55:56,599 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:55:56,613 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:55:56,710 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 776
2026-03-07 05:55:56,710 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:55:56,731 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:55:56,731 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:55:58,230 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:55:58,380 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:55:59,241 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:55:59,243 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 05:56:01,466 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:56:01,466 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:03,671 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token256_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token256_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token256_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:56:03,942 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token896.sima
2026-03-07 05:56:03,952 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:56:03,952 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token896"
2026-03-07 05:56:03,954 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:56:03,954 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:56:03,955 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:56:03,955 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:56:03,955 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:56:03,955 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:56:07,017 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 718
2026-03-07 05:56:07,017 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:56:07,191 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:56:07,433 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:56:07,465 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:56:08,089 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:56:08,090 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:56:08,101 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:56:08,380 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:56:09,163 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:56:09,703 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:56:09,748 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:56:09,776 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.292
2026-03-07 05:56:09,776 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:56:09,782 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:56:09,923 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:56:09,923 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:56:11,527 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:56:12,371 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:56:12,948 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:56:12,949 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:56:12,971 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:56:13,024 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:56:13,054 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.27
2026-03-07 05:56:13,054 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:56:13,060 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:56:13,220 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:56:13,220 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:13,490 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_layer28_conv_mpk.json, llima-compile
2026-03-07 05:56:13,760 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024.sima
2026-03-07 05:56:13,772 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:56:13,772 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024"
2026-03-07 05:56:13,776 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:56:13,776 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:56:13,777 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:56:13,777 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:56:13,777 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:56:13,777 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:56:16,631 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:56:16,632 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 3s.
2026-03-07 05:56:17,284 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 851
2026-03-07 05:56:17,284 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:56:17,656 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:56:17,835 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:56:17,867 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:18,350 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token384_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token384_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token384_mpk.json, llima-compile
2026-03-07 05:56:18,561 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:56:18,565 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:56:18,588 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:56:18,989 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152.sima
2026-03-07 05:56:19,001 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:56:19,001 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152"
2026-03-07 05:56:19,003 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:56:19,003 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:56:19,004 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:56:19,004 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:56:19,004 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:56:19,004 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:56:19,545 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:56:19,725 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:56:22,053 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 971
2026-03-07 05:56:22,054 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:22,585 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token512_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token512_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token512_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:56:22,853 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280.sima
2026-03-07 05:56:22,864 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:56:22,864 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280"
2026-03-07 05:56:22,866 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:56:22,866 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:56:22,867 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:56:22,867 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:56:22,867 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:56:22,867 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:56:25,513 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 750
2026-03-07 05:56:25,513 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:56:30,824 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:56:30,999 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:56:31,039 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:56:31,797 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:56:31,798 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:56:31,817 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:56:34,721 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:56:35,635 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:56:35,687 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:56:35,823 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:56:35,863 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:56:36,301 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:56:36,354 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:56:36,415 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:56:36,449 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.258
2026-03-07 05:56:36,449 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:56:36,456 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:56:36,479 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:56:36,643 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:56:36,643 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:56:36,681 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:56:36,682 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:56:36,711 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:56:39,245 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:56:39,358 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:56:39,398 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:56:40,295 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:56:40,295 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:56:40,328 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:56:40,621 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:56:40,622 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 4s.
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:56:47,562 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token640_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token640_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token640_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:56:47,838 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408.sima
2026-03-07 05:56:47,848 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:56:47,848 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408"
2026-03-07 05:56:47,850 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:56:47,850 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:56:47,851 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:56:47,851 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:56:47,851 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:56:47,851 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:56:47,899 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:56:48,076 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:56:50,380 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 752
2026-03-07 05:56:50,380 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:56:50,829 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:56:51,689 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:56:52,361 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:56:52,420 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:56:52,470 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.246
2026-03-07 05:56:52,470 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:56:52,476 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:56:52,654 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:56:52,655 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:56:56,507 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:56:56,508 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 5s.
2026-03-07 05:57:03,165 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:57:03,269 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:03,270 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token768_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token768_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token768_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:57:03,541 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536.sima
2026-03-07 05:57:03,921 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:57:03,921 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536"
2026-03-07 05:57:03,922 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:57:03,922 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:57:03,924 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:57:03,924 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:57:03,924 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:57:03,924 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:57:04,066 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:57:04,773 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:57:04,835 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:57:04,876 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.241
2026-03-07 05:57:04,876 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:57:04,883 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:57:05,070 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:57:05,071 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:57:06,405 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 833
2026-03-07 05:57:06,406 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:57:09,018 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:57:09,020 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 6s.
2026-03-07 05:57:09,081 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:57:09,319 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:57:09,378 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:57:10,359 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:57:10,360 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:57:10,430 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:57:11,453 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:57:11,702 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:16,368 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token896_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token896_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token896_stage1_mla.elf, llima-compile
2026-03-07 05:57:16,642 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664.sima
2026-03-07 05:57:16,653 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:57:16,653 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664"
2026-03-07 05:57:16,654 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:57:16,654 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:57:16,656 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:57:16,656 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:57:16,656 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:57:16,656 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:57:19,187 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 775
2026-03-07 05:57:19,187 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:57:20,836 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:57:21,171 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:57:25,909 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:57:26,232 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:57:29,939 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:57:30,266 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:57:30,352 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:57:31,410 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:57:31,411 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:57:31,512 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:57:32,893 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:57:34,115 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:57:35,114 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:57:35,195 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:57:35,239 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.233
2026-03-07 05:57:35,239 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:57:35,252 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:57:35,496 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:57:35,496 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:57:40,685 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:57:40,692 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 7s.
2026-03-07 05:57:42,449 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:57:42,759 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:57:42,847 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:57:43,968 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:57:43,969 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:57:44,077 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:57:48,229 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:57:48,582 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:57:48,849 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:57:49,458 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1024_mpk.json, llima-compile
2026-03-07 05:57:49,730 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792.sima
2026-03-07 05:57:49,740 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:57:49,740 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792"
2026-03-07 05:57:49,741 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:57:49,741 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:57:49,742 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:57:49,743 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:57:49,743 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:57:49,743 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:57:50,408 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:57:51,758 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:57:51,855 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 921
2026-03-07 05:57:51,855 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:57:51,859 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:57:51,917 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.23
2026-03-07 05:57:51,917 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:57:51,937 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:57:52,232 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:57:52,232 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:57:55,194 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:57:56,747 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:57:58,112 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:57:58,225 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:57:58,278 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.225
2026-03-07 05:57:58,278 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:57:58,299 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:57:58,636 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:57:58,637 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:57:58,643 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:57:58,649 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 8s.
2026-03-07 05:58:05,674 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:58:05,680 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 8s.
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:09,224 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1152_mpk.json, llima-compile
2026-03-07 05:58:09,496 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920.sima
2026-03-07 05:58:09,506 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:58:09,506 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920"
2026-03-07 05:58:09,508 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:58:09,508 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:58:09,509 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:58:09,509 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:58:09,509 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:58:09,509 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:58:12,459 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 1041
2026-03-07 05:58:12,460 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 05:58:15,561 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:58:15,834 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:58:15,920 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:58:17,086 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:58:17,087 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:17,154 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1280_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:58:17,165 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:58:17,435 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token127.sima
2026-03-07 05:58:17,448 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:58:17,449 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token127"
2026-03-07 05:58:17,450 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:58:17,450 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:58:17,451 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:58:17,452 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:58:17,452 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:58:17,452 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:58:18,647 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:58:18,774 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:58:18,786 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:58:18,862 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:58:18,863 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:58:18,868 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:58:19,319 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:58:20,962 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:58:21,292 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:58:21,328 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:58:22,769 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:58:22,902 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:58:22,958 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.223
2026-03-07 05:58:22,958 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:58:23,003 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:58:23,398 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:58:23,398 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:58:27,381 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:58:27,428 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:58:31,404 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:58:31,663 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:58:31,672 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 10s.
2026-03-07 05:58:31,819 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:58:31,991 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:58:32,002 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:58:32,006 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:58:32,006 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:58:32,006 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:58:32,043 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:58:32,043 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:58:32,766 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:58:36,415 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:58:36,815 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:58:37,229 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 05:58:37,739 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:58:37,836 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:58:38,101 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:58:39,094 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:58:39,095 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:58:39,168 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:39,169 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token127_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token127_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token127_mpk.json, llima-compile
2026-03-07 05:58:39,222 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:58:39,456 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token255.sima
2026-03-07 05:58:39,470 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:58:39,470 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token255"
2026-03-07 05:58:39,472 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:58:39,472 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:58:39,473 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:58:39,473 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:58:39,474 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:58:39,474 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:58:41,203 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:58:41,338 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:58:41,350 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:58:41,494 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:58:41,495 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:58:41,501 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:58:44,865 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1408_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:58:45,148 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token383.sima
2026-03-07 05:58:45,158 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:58:45,158 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token383"
2026-03-07 05:58:45,160 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:58:45,160 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:58:45,161 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:58:45,161 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:58:45,161 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:58:45,161 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:58:47,011 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:58:47,137 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:58:47,149 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:58:47,355 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:58:47,356 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:58:47,361 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:58:51,561 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:58:54,360 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:58:55,127 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:58:55,220 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:58:58,245 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:58:58,377 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:58:58,435 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.219
2026-03-07 05:58:58,435 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:58:58,510 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:58:58,888 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:58:58,889 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:58:59,959 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:00,030 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:03,019 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:03,567 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:03,913 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:03,944 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:03,947 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:59:03,947 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:03,947 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:04,041 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:04,041 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:06,544 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:06,560 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:06,561 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:59:06,746 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:06,761 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 11s.
2026-03-07 05:59:07,077 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:07,317 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:07,351 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:07,376 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:07,379 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:59:07,379 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:07,379 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:07,461 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:07,461 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:07,692 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:09,098 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:09,188 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:09,189 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:09,467 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token255_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token255_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token255_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:59:09,745 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token511.sima
2026-03-07 05:59:09,755 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:09,755 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token511"
2026-03-07 05:59:09,756 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:09,757 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:09,758 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:09,758 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:09,758 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:09,758 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:11,672 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:11,803 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:11,815 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:12,088 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:12,090 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:12,097 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:12,114 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:12,147 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:12,147 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:12,148 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token383_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token383_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token383_stage1_mla.elf, llima-compile
2026-03-07 05:59:12,430 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token639.sima
2026-03-07 05:59:12,441 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:12,441 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token639"
2026-03-07 05:59:12,443 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:12,443 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:12,444 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:12,444 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:12,444 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:12,444 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:13,773 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:13,910 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:13,929 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:13,972 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.218
2026-03-07 05:59:13,972 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:14,042 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:14,053 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:14,054 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:14,392 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:14,393 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:14,398 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:14,455 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:14,455 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:19,665 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1536_mpk.json, llima-compile
2026-03-07 05:59:19,954 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token767.sima
2026-03-07 05:59:19,965 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:19,965 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token767"
2026-03-07 05:59:19,967 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:19,967 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:19,968 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:19,968 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:19,968 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:19,968 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:21,088 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:21,204 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:21,217 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:21,617 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:21,618 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:21,624 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:22,845 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:22,859 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 12s.
2026-03-07 05:59:26,228 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:26,323 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:29,679 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:29,761 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:32,608 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:33,010 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:34,652 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:35,256 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:35,621 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:35,652 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:35,655 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:59:35,655 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:35,656 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:35,752 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:35,753 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:36,713 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:36,714 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1664_stage1_mla.elf, llima-compile
2026-03-07 05:59:36,994 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:37,000 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token895.sima
2026-03-07 05:59:37,010 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:37,010 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token895"
2026-03-07 05:59:37,011 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:37,011 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:37,012 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:37,013 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:37,013 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:37,013 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:37,533 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:37,760 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:37,762 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:59:37,814 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:37,845 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:37,872 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:37,874 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:59:37,875 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:37,875 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:37,913 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:37,962 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:37,962 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:38,442 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:38,951 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:39,049 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:39,062 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:39,529 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:39,530 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:39,536 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:39,803 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:39,804 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:40,960 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token511_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token511_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token511_mpk.json, llima-compile
2026-03-07 05:59:41,238 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023.sima
2026-03-07 05:59:41,251 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:41,251 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023"
2026-03-07 05:59:41,252 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:41,253 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:41,254 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:41,254 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:41,254 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:41,254 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:41,804 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:42,812 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token639_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token639_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token639_mpk.json, llima-compile
2026-03-07 05:59:43,095 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151.sima
2026-03-07 05:59:43,105 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:43,105 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151"
2026-03-07 05:59:43,106 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:43,106 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:43,107 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:43,108 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:43,108 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:43,108 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:43,187 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:43,284 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:43,296 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:43,347 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:43,478 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:43,544 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.215
2026-03-07 05:59:43,544 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:43,642 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:43,827 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:43,827 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:43,834 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:44,030 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:44,030 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:44,641 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:44,725 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:44,737 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:45,339 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:45,340 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:45,345 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:46,370 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 05:59:46,980 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 05:59:47,367 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 05:59:47,398 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 05:59:47,401 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 05:59:47,401 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 05:59:47,401 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 05:59:47,500 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 05:59:47,501 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 05:59:49,600 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:49,602 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 05:59:52,249 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 05:59:52,259 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 13s.
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 05:59:53,140 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 05:59:53,141 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token767_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token767_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token767_stage1_mla_stats.yaml, llima-compile
2026-03-07 05:59:53,435 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279.sima
2026-03-07 05:59:53,444 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 05:59:53,444 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279"
2026-03-07 05:59:53,446 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 05:59:53,446 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 05:59:53,447 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 05:59:53,448 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 05:59:53,448 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 05:59:53,448 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 05:59:54,605 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:54,685 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 05:59:55,417 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 05:59:55,509 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 05:59:55,522 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 05:59:56,183 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 05:59:56,184 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 05:59:56,189 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 05:59:59,801 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 05:59:59,900 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:01,925 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:02,434 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:02,454 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:02,519 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:02,761 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:02,788 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:02,790 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:02,790 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:02,791 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:02,877 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:02,877 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:04,696 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:04,698 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:05,629 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1792_mpk.json, llima-compile
2026-03-07 06:00:05,912 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407.sima
2026-03-07 06:00:05,922 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:05,922 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407"
2026-03-07 06:00:05,923 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:05,923 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:05,924 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:05,925 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:05,925 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:05,925 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:05,937 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:07,514 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 06:00:07,590 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:07,602 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:07,874 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token895_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token895_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token895_mpk.json, llima-compile
2026-03-07 06:00:08,158 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535.sima
2026-03-07 06:00:08,168 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:08,168 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535"
2026-03-07 06:00:08,169 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:08,169 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:08,170 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:08,171 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:08,171 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:08,171 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:08,325 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:08,325 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:08,329 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:08,331 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:08,983 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:09,363 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:09,395 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:09,398 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:09,398 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:09,399 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:09,498 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:09,499 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:09,646 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:09,719 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 06:00:09,793 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:09,806 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:09,952 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:10,512 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:10,597 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:10,598 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:10,605 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:10,844 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:10,872 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:10,875 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:10,875 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:10,875 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:10,965 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:10,965 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:11,613 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:11,614 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 0s.
2026-03-07 06:00:12,849 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:12,850 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:00:13,862 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:14,013 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:14,082 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.214
2026-03-07 06:00:14,082 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:14,186 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:14,614 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:14,614 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:14,743 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:14,841 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:15,260 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:15,261 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:15,261 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:15,261 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:15,261 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1023_mpk.json, llima-compile
2026-03-07 06:00:15,904 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663.sima
2026-03-07 06:00:15,914 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:15,914 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663"
2026-03-07 06:00:15,915 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:15,915 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:15,916 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:15,917 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:15,917 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:15,917 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:16,257 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1151_mpk.json, llima-compile
2026-03-07 06:00:16,539 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791.sima
2026-03-07 06:00:16,549 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:16,549 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791"
2026-03-07 06:00:16,550 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:16,550 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:16,551 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:16,552 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:16,552 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:16,552 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:17,030 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 896
2026-03-07 06:00:17,030 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 06:00:17,698 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 912
2026-03-07 06:00:17,699 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 06:00:22,799 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:23,352 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:23,366 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 14s.
2026-03-07 06:00:23,424 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:23,814 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:23,846 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:23,849 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:23,849 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:23,849 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:23,948 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:23,948 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:24,678 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:24,759 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:25,289 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 06:00:25,517 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:25,550 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:25,999 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 06:00:26,059 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:26,061 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:00:26,238 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:26,271 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:26,436 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:26,437 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:26,450 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:27,228 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:27,228 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:27,241 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:28,729 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:28,819 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:30,249 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1279_mpk.json, llima-compile
2026-03-07 06:00:30,899 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919.sima
2026-03-07 06:00:30,909 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:30,909 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919"
2026-03-07 06:00:30,910 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:30,910 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:30,911 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:30,911 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:30,912 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:30,912 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:32,059 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 1013
2026-03-07 06:00:32,060 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 06:00:32,358 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:32,885 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:33,213 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:33,247 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:33,249 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:33,249 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:33,249 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:33,364 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:33,364 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:35,854 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:35,855 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:00:36,389 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:00:37,010 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:00:37,394 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:00:37,431 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:00:37,434 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:00:37,434 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:00:37,435 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:00:37,558 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:00:37,558 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - 	EV74: 4
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:38,443 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920_mpk.json, models--LiquidAI--LFM2-2.6B_language_n128_cache_token1920_stage1_mla_stats.yaml, llima-compile
2026-03-07 06:00:38,724 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047.sima
2026-03-07 06:00:38,737 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:38,737 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047"
2026-03-07 06:00:38,738 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU
2026-03-07 06:00:38,739 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA
2026-03-07 06:00:38,740 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:38,740 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:38,740 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:38,740 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:39,865 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 06:00:39,921 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/slice_concat_0, 1196
2026-03-07 06:00:39,921 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension
2026-03-07 06:00:40,094 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:40,128 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:40,258 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:00:40,259 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:40,314 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1407_stage1_mla.elf, llima-compile
2026-03-07 06:00:41,047 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-2.6B/sima_files/sdk/models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final.sima
2026-03-07 06:00:41,145 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:41,146 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:41,158 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:41,216 - afe.apis.model - WARNING - A deepcopy is disabled hence model graph will be mutated and no other APIs can be called after the compile step.
2026-03-07 06:00:41,216 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final"
2026-03-07 06:00:41,218 - afe.core.compile_networks - INFO - The model is split into 2 segments for MLA and APU
2026-03-07 06:00:41,218 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 2: compiling for MLA
2026-03-07 06:00:41,292 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties
2026-03-07 06:00:41,293 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting
2026-03-07 06:00:41,293 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts
2026-03-07 06:00:41,293 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters
2026-03-07 06:00:41,334 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts
2026-03-07 06:00:41,341 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:41,354 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:41,874 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:41,875 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:41,882 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:45,228 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:00:45,228 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:00:45,228 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:45,228 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:00:45,229 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1535_stage1_mla.elf, llima-compile
2026-03-07 06:00:47,383 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts
2026-03-07 06:00:47,626 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors
2026-03-07 06:00:47,659 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done
2026-03-07 06:00:48,740 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file
2026-03-07 06:00:48,741 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE
2026-03-07 06:00:48,754 - mlc.compiler.model_graph.l1_based - INFO - Generating model code
2026-03-07 06:00:49,944 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:50,080 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:52,718 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:52,853 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:00:59,435 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:00:59,521 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:01:01,295 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:01:01,987 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:01:02,529 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:01:02,576 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:01:02,578 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:01:02,579 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:01:02,579 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:01:02,723 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:01:02,724 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:01:04,054 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:01:04,737 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:01:05,296 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:01:05,343 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:01:05,346 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:01:05,346 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:01:05,346 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:01:05,494 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:01:05,495 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:01:05,641 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:01:05,738 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:01:05,739 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:01:05,778 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:01:06,912 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:01:08,569 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:01:08,570 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:01:08,912 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:01:09,368 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:01:09,452 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:11,014 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1663_stage1_mla_stats.yaml, llima-compile
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:01:13,853 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:13,854 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1791_stage1_mla_stats.yaml, llima-compile
2026-03-07 06:01:14,219 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done
2026-03-07 06:01:14,355 - mlc.test_util.test_context - INFO - Scheduling instructions
2026-03-07 06:01:17,101 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:01:17,768 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:01:18,324 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:01:18,371 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:01:18,373 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:01:18,373 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:01:18,374 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:01:18,518 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:01:18,518 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:01:21,529 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:01:21,532 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:01:25,478 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization
2026-03-07 06:01:26,229 - mlc.test_util.test_context - INFO - Inserting and merging NOPs
2026-03-07 06:01:26,776 - mlc.test_util.test_context - INFO - Setting IQ sync bits
2026-03-07 06:01:26,824 - mlc.test_util.test_context - INFO - Run compression
2026-03-07 06:01:26,827 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404
2026-03-07 06:01:26,827 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:01:26,828 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:01:26,924 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:26,925 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token1919_stage1_mla_stats.yaml, llima-compile
2026-03-07 06:01:26,974 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:01:26,974 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:01:30,021 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:01:30,022 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 1s.
2026-03-07 06:01:32,244 - mlc.test_util.test_context - INFO - Compression done in 22s. Compression ratio: 0.975
2026-03-07 06:01:32,244 - mlc.test_util.test_context - INFO - Re-allocate dram memory
2026-03-07 06:01:32,924 - mlc.test_util.test_context - INFO - Generating metrics
2026-03-07 06:01:33,206 - mlc.test_util.test_context - INFO - Writing report to MLC file
2026-03-07 06:01:33,206 - mlc.test_util.test_context - INFO - Writing instructions to MLC file
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - 	EV74: 5
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - 	A65 : 0
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:01:35,481 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047_mpk.json, models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_cache_token2047_stage1_mla_stats.yaml, llima-compile
2026-03-07 06:02:29,550 - mlc.test_util.test_context - INFO - Code generation done
2026-03-07 06:02:29,578 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO -   Evaluate model and writing chk file done in 2s.
2026-03-07 06:02:33,097 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 2 of 2: backend is APU
2026-03-07 06:02:33,099 - afe.core.compile_networks - INFO - Stage 2 of 2: compiling for APU
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - ==============================
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - Compilation summary:
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - Desired batch size: 1
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - Achieved batch size: 1
2026-03-07 06:02:44,256 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - Plugin distribution per backend:
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - 	MLA : 1
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - 	EV74: 2
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - 	A65 : 1
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - ------------------------------
2026-03-07 06:02:44,257 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final_stage1_mla_stats.yaml, models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final_stage2_a65.so, models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final_stage1_mla.elf, models--LiquidAI--LFM2-2.6B_language_n1_post_layer29_conv_final_mpk.json, llima-compile
2026-03-07 06:02:46,539 - sima_lmm.model.vision_language_model - INFO - FileGenMode.MODEL_SDK_COMPILE files generation completed.
Generated all mode=MODEL_SDK_COMPILE files