| 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_pre_8 |
| group_post_8 |
| single_pre_8 |
| single_post_8 |
| group_conv_9 |
| single_conv_9 |
| group_pre_10 |
| group_post_10 |
| single_pre_10 |
| single_post_10 |
| group_conv_11 |
| single_conv_11 |
| group_pre_12 |
| group_post_12 |
| single_pre_12 |
| single_post_12 |
| group_conv_13 |
| single_conv_13 |
| group_pre_14 |
| group_post_14 |
| single_pre_14 |
| single_post_14 |
| group_conv_15 |
| single_conv_15 |
| 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_15 |
| vision_0 |
|
|
| 2026-03-07 06:03:32,220 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.DEVKIT files... |
| Generated all mode=DEVKIT files |
| 2026-03-07 06:03:33,670 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.SOURCE_TO_ONNX files... |
| 2026-03-07 06:03:58,597 - sima_lmm.model.vision_language_model - INFO - FileGenMode.SOURCE_TO_ONNX files generation completed. |
| Generated all mode=SOURCE_TO_ONNX files |
| 2026-03-07 06:03:58,597 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.ONNX_TO_QUANT files... |
| Running Calibration ... |
|
|
| Calibration Progress: |██████████████████████████████| 100.0% 1|1 Complete. 1/1 |
| Running Calibration ...DONE |
| Running quantization ... |
| Running quantization ...DONE |
| 2026-03-07 06:06:31,183 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv.onnx'] in onnx format |
| 2026-03-07 06:06:31,183 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv.onnx'] in onnx format |
| 2026-03-07 06:06:31,183 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv.onnx'] in onnx format |
| 2026-03-07 06:06:31,183 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv.onnx'] in onnx format |
| 2026-03-07 06:06:31,184 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2.onnx'] in onnx format |
| 2026-03-07 06:06:31,264 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2.onnx'] in onnx format |
| 2026-03-07 06:06:32,080 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,080 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:32,101 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,101 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:32,174 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,174 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:32,187 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,187 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:32,374 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,374 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:32,417 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:32,417 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:37,083 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:37,122 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:37,949 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:39,850 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv.sima |
| 2026-03-07 06:06:39,853 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2.sima |
| 2026-03-07 06:06:39,854 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2.onnx'] in onnx format |
| 2026-03-07 06:06:39,857 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2.onnx'] in onnx format |
| 2026-03-07 06:06:40,120 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv.sima |
| 2026-03-07 06:06:40,126 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv.onnx'] in onnx format |
| 2026-03-07 06:06:40,302 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:40,303 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:40,419 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:40,818 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:40,818 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:41,056 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:41,056 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:42,037 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:42,695 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:46,404 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:47,010 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2.sima |
| 2026-03-07 06:06:47,013 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv.onnx'] in onnx format |
| 2026-03-07 06:06:47,750 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:06:48,155 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:06:48,202 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:48,202 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:48,977 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2.sima |
| 2026-03-07 06:06:48,983 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv.onnx'] in onnx format |
| 2026-03-07 06:06:49,014 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv.sima |
| 2026-03-07 06:06:49,018 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv.onnx'] in onnx format |
| 2026-03-07 06:06:49,166 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv.sima |
| 2026-03-07 06:06:49,170 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5.onnx'] in onnx format |
| 2026-03-07 06:06:49,403 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:49,709 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:49,709 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:50,045 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:50,045 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:50,150 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:50,150 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:50,333 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2.sima |
| 2026-03-07 06:06:50,345 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5.onnx'] in onnx format |
| 2026-03-07 06:06:51,084 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:51,084 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:53,240 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:53,915 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:54,873 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:55,661 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv.sima |
| 2026-03-07 06:06:55,671 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5.onnx'] in onnx format |
| 2026-03-07 06:06:55,966 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:06:57,757 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:57,758 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:57,880 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv.sima |
| 2026-03-07 06:06:57,885 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5.onnx'] in onnx format |
| 2026-03-07 06:06:58,391 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5.sima |
| 2026-03-07 06:06:58,394 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv.onnx'] in onnx format |
| 2026-03-07 06:06:58,418 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:58,418 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:58,818 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv.sima |
| 2026-03-07 06:06:58,823 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv.onnx'] in onnx format |
| 2026-03-07 06:06:59,365 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:59,365 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:59,598 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:06:59,598 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:06:59,946 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:01,032 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:02,530 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:03,945 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5.sima |
| 2026-03-07 06:07:03,947 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv.onnx'] in onnx format |
| 2026-03-07 06:07:04,010 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:04,987 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5.sima |
| 2026-03-07 06:07:04,999 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv.onnx'] in onnx format |
| 2026-03-07 06:07:05,387 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5.sima |
| 2026-03-07 06:07:05,391 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8.onnx'] in onnx format |
| 2026-03-07 06:07:05,417 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:05,417 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:05,702 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:05,702 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:06,155 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv.sima |
| 2026-03-07 06:07:06,166 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8.onnx'] in onnx format |
| 2026-03-07 06:07:06,347 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:06,347 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:06,348 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:07,007 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:07,008 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:08,667 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv.sima |
| 2026-03-07 06:07:08,671 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8.onnx'] in onnx format |
| 2026-03-07 06:07:09,614 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:09,614 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:10,715 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:10,732 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:12,090 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:14,413 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv.sima |
| 2026-03-07 06:07:14,419 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8.onnx'] in onnx format |
| 2026-03-07 06:07:14,907 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:14,907 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:14,980 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8.sima |
| 2026-03-07 06:07:14,993 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv.onnx'] in onnx format |
| 2026-03-07 06:07:15,776 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv.sima |
| 2026-03-07 06:07:15,788 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv.onnx'] in onnx format |
| 2026-03-07 06:07:15,839 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:16,024 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:16,024 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:16,250 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:16,712 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:16,712 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:19,572 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:20,495 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:20,566 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8.sima |
| 2026-03-07 06:07:20,569 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10.onnx'] in onnx format |
| 2026-03-07 06:07:20,894 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8.sima |
| 2026-03-07 06:07:20,900 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10.onnx'] in onnx format |
| 2026-03-07 06:07:21,549 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv.sima |
| 2026-03-07 06:07:21,558 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10.onnx'] in onnx format |
| 2026-03-07 06:07:21,626 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:21,626 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:21,659 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:21,660 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:22,131 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8.sima |
| 2026-03-07 06:07:22,135 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10.onnx'] in onnx format |
| 2026-03-07 06:07:22,521 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:22,521 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:22,624 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:22,624 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:23,522 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:25,786 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv.sima |
| 2026-03-07 06:07:25,789 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv.onnx'] in onnx format |
| 2026-03-07 06:07:26,311 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:26,738 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:26,738 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:26,923 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:27,505 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:28,068 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:28,636 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10.sima |
| 2026-03-07 06:07:28,639 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv.onnx'] in onnx format |
| 2026-03-07 06:07:29,871 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:29,871 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:29,949 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10.sima |
| 2026-03-07 06:07:29,958 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12.onnx'] in onnx format |
| 2026-03-07 06:07:30,789 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10.sima |
| 2026-03-07 06:07:30,794 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12.onnx'] in onnx format |
| 2026-03-07 06:07:30,908 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:31,060 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:31,063 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:31,702 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:31,702 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:33,290 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv.sima |
| 2026-03-07 06:07:33,296 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12.onnx'] in onnx format |
| 2026-03-07 06:07:34,134 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10.sima |
| 2026-03-07 06:07:34,139 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12.onnx'] in onnx format |
| 2026-03-07 06:07:34,290 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:34,290 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:34,613 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:34,613 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:35,374 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:37,053 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:37,686 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv.sima |
| 2026-03-07 06:07:37,694 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv.onnx'] in onnx format |
| 2026-03-07 06:07:38,359 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:38,588 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:39,084 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12.sima |
| 2026-03-07 06:07:39,089 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv.onnx'] in onnx format |
| 2026-03-07 06:07:39,330 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:39,331 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:39,802 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:39,802 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:40,112 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:40,531 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12.sima |
| 2026-03-07 06:07:40,541 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14.onnx'] in onnx format |
| 2026-03-07 06:07:40,666 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:42,681 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12.sima |
| 2026-03-07 06:07:42,684 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14.onnx'] in onnx format |
| 2026-03-07 06:07:42,880 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:42,880 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:43,557 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:43,557 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:44,264 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv.sima |
| 2026-03-07 06:07:44,269 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14.onnx'] in onnx format |
| 2026-03-07 06:07:44,434 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12.sima |
| 2026-03-07 06:07:44,437 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14.onnx'] in onnx format |
| 2026-03-07 06:07:44,841 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:44,841 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:45,214 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:45,214 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:46,536 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:48,485 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:48,573 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:48,818 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv.sima |
| 2026-03-07 06:07:48,821 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv.onnx'] in onnx format |
| 2026-03-07 06:07:49,129 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:49,129 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:50,403 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14.sima |
| 2026-03-07 06:07:50,409 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv.onnx'] in onnx format |
| 2026-03-07 06:07:50,596 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:50,596 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:50,791 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:51,067 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14.sima |
| 2026-03-07 06:07:51,077 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0.onnx'] in onnx format |
| 2026-03-07 06:07:51,229 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:51,230 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:51,524 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:52,067 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:53,413 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14.sima |
| 2026-03-07 06:07:53,419 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128.onnx'] in onnx format |
| 2026-03-07 06:07:53,591 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:53,591 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:53,925 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:55,431 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:56,335 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:56,908 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv.sima |
| 2026-03-07 06:07:56,912 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256.onnx'] in onnx format |
| 2026-03-07 06:07:56,994 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:56,994 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:57,017 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv.sima |
| 2026-03-07 06:07:57,020 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384.onnx'] in onnx format |
| 2026-03-07 06:07:57,062 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv.sima |
| 2026-03-07 06:07:57,067 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512.onnx'] in onnx format |
| 2026-03-07 06:07:57,157 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:57,157 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:57,176 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:57,176 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:57,415 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14.sima |
| 2026-03-07 06:07:57,418 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640.onnx'] in onnx format |
| 2026-03-07 06:07:57,533 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:57,533 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:57,645 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0.sima |
| 2026-03-07 06:07:57,649 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768.onnx'] in onnx format |
| 2026-03-07 06:07:57,731 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:57,731 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:07:57,825 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:59,693 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128.sima |
| 2026-03-07 06:07:59,695 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896.onnx'] in onnx format |
| 2026-03-07 06:07:59,712 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:07:59,842 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:07:59,842 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:00,078 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:00,575 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:00,842 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:01,370 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:01,884 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256.sima |
| 2026-03-07 06:08:01,887 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024.onnx'] in onnx format |
| 2026-03-07 06:08:02,005 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:02,005 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:03,232 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:04,066 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384.sima |
| 2026-03-07 06:08:04,071 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152.onnx'] in onnx format |
| 2026-03-07 06:08:04,108 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512.sima |
| 2026-03-07 06:08:04,112 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280.onnx'] in onnx format |
| 2026-03-07 06:08:04,158 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:04,158 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:04,196 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:04,196 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:04,711 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640.sima |
| 2026-03-07 06:08:04,713 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408.onnx'] in onnx format |
| 2026-03-07 06:08:04,799 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:04,800 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:05,207 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768.sima |
| 2026-03-07 06:08:05,213 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536.onnx'] in onnx format |
| 2026-03-07 06:08:05,297 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:05,297 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:05,441 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:07,024 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896.sima |
| 2026-03-07 06:08:07,027 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664.onnx'] in onnx format |
| 2026-03-07 06:08:07,031 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:07,130 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:07,131 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:07,144 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:07,642 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:08,725 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:10,050 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024.sima |
| 2026-03-07 06:08:10,053 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792.onnx'] in onnx format |
| 2026-03-07 06:08:10,142 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:10,143 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:11,027 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:11,946 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152.sima |
| 2026-03-07 06:08:11,951 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920.onnx'] in onnx format |
| 2026-03-07 06:08:12,036 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:12,036 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:12,459 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:12,943 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280.sima |
| 2026-03-07 06:08:12,946 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127.onnx'] in onnx format |
| 2026-03-07 06:08:13,029 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:13,029 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:13,133 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:13,202 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:13,277 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408.sima |
| 2026-03-07 06:08:13,281 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255.onnx'] in onnx format |
| 2026-03-07 06:08:13,388 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:13,388 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:14,072 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:14,228 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536.sima |
| 2026-03-07 06:08:14,231 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383.onnx'] in onnx format |
| 2026-03-07 06:08:14,315 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:14,315 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:15,124 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:16,005 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:16,714 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:17,100 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:17,100 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127.sima |
| 2026-03-07 06:08:17,105 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511.onnx'] in onnx format |
| 2026-03-07 06:08:17,210 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:17,211 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:17,225 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664.sima |
| 2026-03-07 06:08:17,231 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639.onnx'] in onnx format |
| 2026-03-07 06:08:17,380 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:17,380 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:17,502 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:17,630 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255.sima |
| 2026-03-07 06:08:17,633 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767.onnx'] in onnx format |
| 2026-03-07 06:08:17,808 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:17,808 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:18,324 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383.sima |
| 2026-03-07 06:08:18,327 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895.onnx'] in onnx format |
| 2026-03-07 06:08:18,409 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:18,409 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:19,480 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:19,598 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792.sima |
| 2026-03-07 06:08:19,600 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023.onnx'] in onnx format |
| 2026-03-07 06:08:19,684 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:19,684 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:20,240 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:20,253 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:21,277 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:21,566 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511.sima |
| 2026-03-07 06:08:21,569 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151.onnx'] in onnx format |
| 2026-03-07 06:08:21,598 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639.sima |
| 2026-03-07 06:08:21,600 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279.onnx'] in onnx format |
| 2026-03-07 06:08:21,624 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:21,649 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:21,649 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:21,753 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:21,753 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:22,147 - afe.ir.transform.calibration_transforms - INFO - Calibration progress: completed 1 samples |
| 2026-03-07 06:08:22,324 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920.sima |
| 2026-03-07 06:08:22,327 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407.onnx'] in onnx format |
| 2026-03-07 06:08:22,410 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:22,410 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:22,708 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767.sima |
| 2026-03-07 06:08:22,710 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535.onnx'] in onnx format |
| 2026-03-07 06:08:22,791 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:22,791 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:22,913 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:23,421 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895.sima |
| 2026-03-07 06:08:23,424 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663.onnx'] in onnx format |
| 2026-03-07 06:08:23,506 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:23,506 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:23,980 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023.sima |
| 2026-03-07 06:08:23,983 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791.onnx'] in onnx format |
| 2026-03-07 06:08:24,065 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:24,065 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:24,377 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:25,027 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:25,320 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:25,588 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:25,668 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151.sima |
| 2026-03-07 06:08:25,673 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919.onnx'] in onnx format |
| 2026-03-07 06:08:25,923 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:25,923 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:26,692 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279.sima |
| 2026-03-07 06:08:26,697 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047.onnx'] in onnx format |
| 2026-03-07 06:08:26,778 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:26,780 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:26,780 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:26,951 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:27,357 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407.sima |
| 2026-03-07 06:08:27,360 - afe.apis.loaded_net - INFO - Loading ['CompiledModels/models--LiquidAI--LFM2-VL-450M/onnx_files/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final.onnx'] in onnx format |
| 2026-03-07 06:08:27,554 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535.sima |
| 2026-03-07 06:08:28,521 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663.sima |
| 2026-03-07 06:08:28,626 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791.sima |
| 2026-03-07 06:08:29,122 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:29,613 - afe.apis.loaded_net - INFO - Quantize loaded net, layout = NCHW, arm_only = False |
| 2026-03-07 06:08:29,613 - afe.apis.loaded_net - INFO - Calibration method = mse |
| 2026-03-07 06:08:29,844 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:30,498 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919.sima |
| 2026-03-07 06:08:31,257 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047.sima |
| 2026-03-07 06:08:34,460 - afe.backends.mla.mla_checkers - INFO - Cannot assign node cast_10, source_name(['argmax']) to MLA. |
| ['Unsupported'] |
| 2026-03-07 06:08:34,945 - afe.ir.transform.calibration_transforms - INFO - Running Calibration ... |
| 2026-03-07 06:08:43,186 - afe.ir.serializer.api - INFO - Saved model: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final.sima |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 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 |
| 2026-03-07 06:08:45,475 - sima_lmm.model.vision_language_model - INFO - FileGenMode.ONNX_TO_QUANT files generation completed. |
| Generated all mode=ONNX_TO_QUANT files |
| 2026-03-07 06:08:45,475 - sima_lmm.model.vision_language_model - INFO - Generating FileGenMode.MODEL_SDK_COMPILE files... |
| 2026-03-07 06:08:50,274 - 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 06:19:09,203 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv.sima |
| 2026-03-07 06:19:09,216 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv.sima |
| 2026-03-07 06:19:09,235 - 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 06:19:09,235 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv" |
| 2026-03-07 06:19:09,238 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,239 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,246 - 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 06:19:09,247 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv" |
| 2026-03-07 06:19:09,250 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,250 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,251 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,251 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,251 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,251 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:09,262 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,263 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,263 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,263 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:09,302 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2.sima |
| 2026-03-07 06:19:09,319 - 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 06:19:09,319 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2" |
| 2026-03-07 06:19:09,323 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:09,323 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,323 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,328 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,328 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,328 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,328 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:09,336 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:09,343 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv.sima |
| 2026-03-07 06:19:09,350 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:09,364 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:09,371 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:09,385 - 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 06:19:09,385 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv" |
| 2026-03-07 06:19:09,385 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:09,390 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv.sima |
| 2026-03-07 06:19:09,390 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,390 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,411 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,412 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,412 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,412 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:09,428 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:09,429 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:09,433 - 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 06:19:09,433 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv" |
| 2026-03-07 06:19:09,438 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,438 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,442 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:09,445 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:09,445 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:09,459 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,459 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:09,459 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,459 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,459 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:09,745 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2.sima |
| 2026-03-07 06:19:09,788 - 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 06:19:09,788 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2" |
| 2026-03-07 06:19:09,792 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:09,792 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:09,810 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:09,810 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:09,810 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:09,810 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:12,400 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:12,420 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:12,459 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:12,480 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:14,141 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:14,212 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:14,368 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:14,435 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:14,477 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:14,665 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:14,665 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:14,676 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:14,731 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:14,852 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:14,865 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:14,871 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:14,887 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:14,948 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:14,965 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:15,205 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:15,241 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:15,281 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:15,335 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:15,470 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:15,471 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:15,484 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:15,622 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:15,639 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:15,675 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:15,675 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:15,776 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:15,776 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:15,792 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:15,812 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:15,813 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:15,826 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:16,953 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.98 |
| 2026-03-07 06:19:16,953 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:17,129 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.98 |
| 2026-03-07 06:19:17,129 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:17,544 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:17,603 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:17,603 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:17,712 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:17,771 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:17,772 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:23,259 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:23,343 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:23,482 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:23,686 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:24,782 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:24,919 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:24,925 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:25,037 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:25,400 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:25,513 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:28,401 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:19:28,449 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,272 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer0_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:19:30,306 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:30,337 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer1_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:19:30,511 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2.sima |
| 2026-03-07 06:19:30,531 - 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 06:19:30,531 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2" |
| 2026-03-07 06:19:30,533 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:30,533 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:30,537 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:30,538 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:30,538 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:30,538 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:30,579 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:30,584 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:30,595 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:30,639 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:30,640 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:30,648 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2.sima |
| 2026-03-07 06:19:30,649 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:30,668 - 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 06:19:30,668 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2" |
| 2026-03-07 06:19:30,671 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:30,672 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:30,676 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:30,676 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:30,676 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:30,676 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:30,797 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:30,810 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:30,843 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:30,859 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:30,859 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:30,867 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:30,986 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:31,388 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:31,460 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:32,172 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:32,190 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:32,631 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:32,646 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:33,908 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:33,958 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:34,188 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:34,255 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:34,262 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:34,380 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:34,408 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:34,451 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:34,464 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:34,518 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.966 |
| 2026-03-07 06:19:34,518 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:34,536 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:34,562 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:34,562 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:34,764 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.971 |
| 2026-03-07 06:19:34,764 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:34,832 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:34,921 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:35,066 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:35,066 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:35,307 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:35,601 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:35,811 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:35,818 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:35,893 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:36,141 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.978 |
| 2026-03-07 06:19:36,142 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:36,206 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:36,251 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:36,251 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:36,321 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:36,402 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:37,481 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:38,523 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:39,059 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:39,107 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:39,562 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.956 |
| 2026-03-07 06:19:39,562 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:39,572 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:39,725 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:39,725 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:39,836 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.974 |
| 2026-03-07 06:19:39,836 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:39,920 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:40,190 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:40,190 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:40,303 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.974 |
| 2026-03-07 06:19:40,303 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:40,387 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:40,659 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:40,659 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:41,068 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:43,694 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:43,700 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:19:47,124 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:47,124 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:19:47,560 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:19:48,030 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,063 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer2_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:19:48,524 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv.sima |
| 2026-03-07 06:19:48,563 - 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 06:19:48,563 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv" |
| 2026-03-07 06:19:48,567 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:48,567 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:48,587 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:48,588 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:48,588 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:48,588 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,657 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer2_stage1_mla.elf, llima-compile |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:48,891 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer2_mpk.json, llima-compile |
| 2026-03-07 06:19:48,930 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv.sima |
| 2026-03-07 06:19:48,962 - 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 06:19:48,962 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv" |
| 2026-03-07 06:19:48,965 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:48,965 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:48,971 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:48,972 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:48,972 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:48,972 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:49,045 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:49,073 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:49,094 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:49,153 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:49,154 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:49,167 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:49,731 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv.sima |
| 2026-03-07 06:19:49,766 - 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 06:19:49,766 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv" |
| 2026-03-07 06:19:49,771 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:49,771 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:49,783 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:49,784 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:49,784 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:49,784 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:52,201 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:52,227 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:19:52,831 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer2_mpk.json, llima-compile |
| 2026-03-07 06:19:53,109 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv.sima |
| 2026-03-07 06:19:53,135 - 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 06:19:53,135 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv" |
| 2026-03-07 06:19:53,138 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:19:53,138 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:19:53,150 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:19:53,150 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:19:53,150 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:19:53,150 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:19:53,221 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:53,249 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:53,269 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:53,326 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:53,327 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:53,339 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:53,732 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:53,957 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:54,153 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:54,207 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:54,339 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:54,340 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:54,353 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:54,461 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:54,468 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:19:54,539 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:54,635 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:54,650 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:19:55,036 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:19:55,043 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:19:55,319 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:19:55,752 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:19:55,808 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:19:55,935 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:19:55,936 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:19:55,949 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:19:56,310 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:19:56,331 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:19:56,724 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.979 |
| 2026-03-07 06:19:56,724 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:19:56,804 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:19:56,863 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:19:56,864 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:19:58,052 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:19:58,628 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:19:58,723 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:19:58,738 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:01,369 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.973 |
| 2026-03-07 06:20:01,369 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:01,450 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:01,510 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:01,510 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:01,835 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer0_conv_mpk.json, llima-compile |
| 2026-03-07 06:20:02,223 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5.sima |
| 2026-03-07 06:20:02,266 - 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 06:20:02,266 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5" |
| 2026-03-07 06:20:02,279 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:02,279 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:02,289 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:02,290 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:02,290 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:02,290 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:02,342 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer1_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:20:02,650 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5.sima |
| 2026-03-07 06:20:02,682 - 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 06:20:02,682 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5" |
| 2026-03-07 06:20:02,686 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:02,686 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:02,696 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:02,696 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:02,696 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:02,697 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:02,760 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:04,185 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:04,298 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:05,802 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:05,917 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:07,338 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:07,408 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:07,426 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:07,450 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:07,630 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:07,630 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:07,639 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:07,800 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:08,202 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:08,239 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:08,340 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:08,341 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:08,356 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:13,113 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:14,008 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:14,534 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:14,616 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:14,678 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:15,167 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:20:15,199 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:20:15,444 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:15,532 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:15,576 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:16,093 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:16,175 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,091 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer4_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:17,119 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer3_conv_mpk.json, llima-compile |
| 2026-03-07 06:20:17,245 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5.sima |
| 2026-03-07 06:20:17,266 - 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 06:20:17,266 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5" |
| 2026-03-07 06:20:17,268 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:17,268 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:17,272 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:17,273 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:17,273 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:17,273 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:17,315 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:17,319 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:17,330 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:17,376 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:17,376 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:17,386 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:17,467 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5.sima |
| 2026-03-07 06:20:17,488 - 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 06:20:17,488 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5" |
| 2026-03-07 06:20:17,491 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:17,492 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:17,496 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:17,497 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:17,497 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:17,497 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:17,618 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:17,631 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:17,664 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:17,679 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:17,680 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:17,688 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:18,206 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:18,345 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:18,593 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.973 |
| 2026-03-07 06:20:18,594 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:18,678 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:18,951 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:18,952 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:18,995 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:19,014 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:19,856 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:19,871 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:20,110 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.964 |
| 2026-03-07 06:20:20,110 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:20,193 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:20,465 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:20,465 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:20,741 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:21,021 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:21,086 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:21,093 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:21,161 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:21,353 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.965 |
| 2026-03-07 06:20:21,353 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:21,370 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:21,396 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:21,396 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:21,581 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:21,651 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:21,663 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:22,439 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:22,486 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:23,173 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:23,329 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.977 |
| 2026-03-07 06:20:23,329 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:23,393 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:23,438 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:23,438 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:23,582 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:23,649 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:26,887 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.97 |
| 2026-03-07 06:20:26,887 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:26,957 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:27,201 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:27,201 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:28,223 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:30,107 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:31,157 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:31,695 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:31,743 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:32,208 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.954 |
| 2026-03-07 06:20:32,208 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:32,228 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:32,382 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:32,383 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:33,450 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:33,450 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:20:33,913 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:20:34,066 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:34,703 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer5_mpk.json, llima-compile |
| 2026-03-07 06:20:34,815 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:34,816 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:20:35,172 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv.sima |
| 2026-03-07 06:20:35,206 - 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 06:20:35,206 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv" |
| 2026-03-07 06:20:35,210 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:35,210 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:35,222 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:35,223 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:35,223 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:35,223 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:35,289 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer5_mpk.json, llima-compile |
| 2026-03-07 06:20:35,578 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv.sima |
| 2026-03-07 06:20:35,604 - 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 06:20:35,604 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv" |
| 2026-03-07 06:20:35,606 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:35,607 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:35,613 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:35,613 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:35,613 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:35,613 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:35,686 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:35,714 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:35,734 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:35,782 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:35,782 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:35,796 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:36,565 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:36,568 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:20:38,808 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:38,829 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:39,379 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:39,380 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:40,377 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer3_conv_mpk.json, llima-compile |
| 2026-03-07 06:20:40,592 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:40,803 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:40,838 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv.sima |
| 2026-03-07 06:20:40,871 - 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 06:20:40,871 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv" |
| 2026-03-07 06:20:40,876 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:40,876 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:40,888 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:40,888 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:40,888 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:40,888 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,088 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer5_mpk.json, llima-compile |
| 2026-03-07 06:20:41,228 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:41,282 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:41,364 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv.sima |
| 2026-03-07 06:20:41,392 - 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 06:20:41,392 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv" |
| 2026-03-07 06:20:41,395 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:41,395 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:41,408 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:41,408 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:41,408 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:41,409 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:41,423 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:41,423 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:41,436 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:41,481 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:41,509 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:41,530 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:41,586 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:41,586 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:41,599 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:41,732 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:41,828 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:41,844 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,897 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:41,898 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:41,898 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:41,898 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:41,898 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:41,898 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer4_conv_mpk.json, llima-compile |
| 2026-03-07 06:20:42,273 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8.sima |
| 2026-03-07 06:20:42,322 - 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 06:20:42,322 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8" |
| 2026-03-07 06:20:42,328 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:42,329 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:42,336 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:42,337 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:42,337 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:42,337 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:43,871 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.979 |
| 2026-03-07 06:20:43,871 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:43,953 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:44,013 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:44,013 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:44,660 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:44,680 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:45,337 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:45,338 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:45,338 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:20:45,338 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:45,338 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:45,338 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer5_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:20:46,068 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8.sima |
| 2026-03-07 06:20:46,098 - 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 06:20:46,098 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8" |
| 2026-03-07 06:20:46,101 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:46,102 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:46,111 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:46,112 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:46,112 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:46,112 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:46,360 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:46,416 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:46,779 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:46,833 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:46,958 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:46,966 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:46,994 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:46,997 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:47,137 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:47,152 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:20:48,215 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:48,616 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:48,652 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:48,751 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:48,751 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:48,766 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:49,227 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.973 |
| 2026-03-07 06:20:49,227 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:20:49,306 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:20:49,366 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:20:49,366 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:20:49,873 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:50,734 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:50,803 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:50,845 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:51,025 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:51,026 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:51,035 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:51,055 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:51,166 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:53,999 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:20:55,275 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:20:55,932 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer6_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:20:56,311 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8.sima |
| 2026-03-07 06:20:56,328 - 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 06:20:56,328 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8" |
| 2026-03-07 06:20:56,331 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:20:56,331 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:20:56,336 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:20:56,336 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:20:56,336 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:20:56,336 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:20:56,458 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:20:56,472 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:20:56,505 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:20:56,521 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:20:56,521 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:20:56,530 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:20:57,017 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:57,129 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:57,853 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:57,872 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:58,214 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:58,354 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:58,870 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:20:58,958 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:20:59,599 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:59,882 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:20:59,905 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:20:59,949 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:20:59,957 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:00,212 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.97 |
| 2026-03-07 06:21:00,212 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:00,227 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:00,254 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:00,255 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:00,405 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:00,805 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:01,324 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:01,324 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:01,406 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:02,338 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer7_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:21:02,476 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8.sima |
| 2026-03-07 06:21:02,497 - 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 06:21:02,497 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8" |
| 2026-03-07 06:21:02,499 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:02,499 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:02,503 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:02,503 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:02,503 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:02,503 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:02,546 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:02,550 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:02,562 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:02,598 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:02,598 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:02,608 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:05,053 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:05,068 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:05,424 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.972 |
| 2026-03-07 06:21:05,424 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:05,508 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:05,783 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:05,784 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:05,954 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:06,352 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:06,632 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:06,773 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:06,843 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:06,856 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:07,023 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:07,977 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:08,067 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:08,137 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:08,485 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:08,566 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:08,599 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.977 |
| 2026-03-07 06:21:08,599 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:08,664 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:08,709 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:08,709 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:10,045 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:11,103 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:11,368 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.969 |
| 2026-03-07 06:21:11,368 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:11,435 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:11,638 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:11,672 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:11,672 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:11,687 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:12,137 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.959 |
| 2026-03-07 06:21:12,138 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:12,147 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:12,300 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:12,301 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:12,500 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.963 |
| 2026-03-07 06:21:12,500 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:12,582 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:12,852 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:12,852 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:13,471 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:16,440 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:16,443 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:16,884 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:17,485 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer8_stage1_mla.elf, llima-compile |
| 2026-03-07 06:21:17,949 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv.sima |
| 2026-03-07 06:21:17,986 - 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 06:21:17,986 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv" |
| 2026-03-07 06:21:17,990 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:17,990 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:18,003 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:18,003 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:18,003 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:18,003 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:20,365 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:20,365 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 6s. |
| 2026-03-07 06:21:20,667 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:20,918 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer8_stage1_mla.elf, llima-compile |
| 2026-03-07 06:21:21,323 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv.sima |
| 2026-03-07 06:21:21,349 - 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 06:21:21,349 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv" |
| 2026-03-07 06:21:21,352 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:21,352 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:21,358 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:21,359 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:21,359 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:21,359 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:21,434 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:21,462 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:21,483 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:21,542 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:21,543 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:21,947 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:21,984 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer8_mpk.json, llima-compile |
| 2026-03-07 06:21:22,381 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10.sima |
| 2026-03-07 06:21:22,401 - 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 06:21:22,401 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10" |
| 2026-03-07 06:21:22,404 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:22,405 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:22,409 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:22,409 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:22,409 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:22,409 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:23,628 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:23,734 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:23,740 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:21:24,049 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:24,104 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:24,238 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:24,239 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:24,252 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:24,939 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:24,960 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:26,697 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:27,134 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:27,135 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:21:27,135 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:27,135 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:27,135 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer6_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:21:27,149 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:27,157 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:21:27,293 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:27,388 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:27,406 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:27,450 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10.sima |
| 2026-03-07 06:21:27,482 - 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 06:21:27,482 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10" |
| 2026-03-07 06:21:27,485 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:27,485 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:27,495 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:27,496 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:27,496 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:27,496 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:27,902 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:28,309 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:28,346 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:28,450 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:28,451 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:28,465 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:29,469 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.97 |
| 2026-03-07 06:21:29,470 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:29,554 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:29,601 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer8_stage1_mla.elf, llima-compile |
| 2026-03-07 06:21:29,617 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:29,617 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:29,982 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10.sima |
| 2026-03-07 06:21:29,999 - 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 06:21:29,999 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10" |
| 2026-03-07 06:21:30,003 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:30,003 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:30,007 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:30,008 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:30,008 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:30,008 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:30,129 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:30,142 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:30,175 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:30,190 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:30,191 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:30,200 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:31,492 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:31,511 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:32,536 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:32,607 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:32,649 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:32,784 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:32,826 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:32,827 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:32,836 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:33,065 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:33,133 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:33,139 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:33,394 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.958 |
| 2026-03-07 06:21:33,394 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:33,409 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:33,436 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:33,436 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:33,832 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:33,941 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:34,176 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:34,177 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer7_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:21:34,301 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10.sima |
| 2026-03-07 06:21:34,326 - 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 06:21:34,327 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10" |
| 2026-03-07 06:21:34,329 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:34,329 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:34,333 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:34,333 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:34,333 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:34,333 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:34,374 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:34,379 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:34,390 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:34,430 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:34,431 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:34,439 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:34,494 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:35,453 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:36,414 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:36,430 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:37,727 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:38,150 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:38,221 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:38,234 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:38,363 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:38,503 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:39,457 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:39,928 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.97 |
| 2026-03-07 06:21:39,928 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:40,016 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:40,064 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:40,064 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:40,811 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:40,899 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:41,363 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:41,364 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer9_conv_mpk.json, llima-compile |
| 2026-03-07 06:21:41,824 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv.sima |
| 2026-03-07 06:21:41,867 - 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 06:21:41,867 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv" |
| 2026-03-07 06:21:41,876 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:41,876 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:41,903 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:41,904 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:41,905 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:41,905 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:42,713 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:43,614 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:44,127 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:44,207 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:44,937 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:47,448 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:47,797 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:47,871 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:47,925 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:48,027 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:21:48,054 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:48,055 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:48,067 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:48,142 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.96 |
| 2026-03-07 06:21:48,142 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:48,225 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:48,473 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:48,495 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:48,496 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:48,585 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:48,586 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer10_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:21:48,864 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv.sima |
| 2026-03-07 06:21:48,877 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:48,891 - 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 06:21:48,891 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv" |
| 2026-03-07 06:21:48,894 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:48,894 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:48,900 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:48,900 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:48,900 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:48,900 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:48,946 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:48,972 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:21:49,000 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:21:49,020 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:21:49,076 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:21:49,077 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:21:49,090 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:21:50,081 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:51,130 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:51,664 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:51,713 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:52,028 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:52,096 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:52,116 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:52,170 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.947 |
| 2026-03-07 06:21:52,170 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:52,189 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:52,215 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.96 |
| 2026-03-07 06:21:52,215 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:52,283 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:52,343 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:52,343 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:52,518 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:52,518 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:21:53,351 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer10_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:21:53,727 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12.sima |
| 2026-03-07 06:21:53,743 - 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 06:21:53,743 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12" |
| 2026-03-07 06:21:53,746 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:21:53,746 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:21:53,751 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:21:53,751 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:21:53,751 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:21:53,751 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:21:54,304 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:21:54,889 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:21:54,985 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:21:55,001 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:21:56,431 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:21:56,434 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:21:57,023 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.975 |
| 2026-03-07 06:21:57,023 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:21:57,106 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:21:57,166 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:21:57,166 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:21:57,441 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:21:57,551 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:21:59,735 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:00,137 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:00,173 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:00,273 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:00,273 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:00,287 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:00,805 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer10_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:22:01,207 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12.sima |
| 2026-03-07 06:22:01,235 - 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 06:22:01,235 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12" |
| 2026-03-07 06:22:01,239 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:01,239 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:01,249 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:01,249 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:01,249 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:01,249 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:02,861 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:02,861 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:22:03,018 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:04,657 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:04,663 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:22:05,818 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:05,889 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:05,932 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:06,116 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:06,116 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:06,126 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:06,881 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:22:06,979 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:07,877 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:08,389 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:08,472 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:08,761 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer11_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:22:09,138 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12.sima |
| 2026-03-07 06:22:09,155 - 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 06:22:09,156 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12" |
| 2026-03-07 06:22:09,159 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:09,159 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:09,164 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:09,164 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:09,164 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:09,164 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:09,286 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:09,299 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:09,332 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:09,332 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:09,348 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:09,348 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:09,358 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:09,471 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:09,737 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer9_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:22:09,870 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12.sima |
| 2026-03-07 06:22:09,890 - 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 06:22:09,890 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12" |
| 2026-03-07 06:22:09,892 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:09,892 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:09,896 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:09,896 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:09,897 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:09,897 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:09,937 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:09,942 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:09,953 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:10,000 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:10,004 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:10,012 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:10,951 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer10_mpk.json, llima-compile |
| 2026-03-07 06:22:11,089 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:11,108 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:11,416 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv.sima |
| 2026-03-07 06:22:11,451 - 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 06:22:11,451 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv" |
| 2026-03-07 06:22:11,455 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:11,456 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:11,468 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:11,469 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:11,469 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:11,469 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:12,374 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.966 |
| 2026-03-07 06:22:12,374 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:12,397 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:12,428 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:12,443 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:12,459 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:12,678 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:12,731 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:12,731 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:12,745 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:12,753 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:13,010 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.959 |
| 2026-03-07 06:22:13,011 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:13,028 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:13,054 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:13,054 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:13,729 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:13,880 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:13,968 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:14,124 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:14,149 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:14,220 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:14,232 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:15,912 - mlc.test_util.test_context - INFO - Compression done in 1s. Compression ratio: 0.98 |
| 2026-03-07 06:22:15,912 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:15,976 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:16,021 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:16,022 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:16,647 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:17,088 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:17,146 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:17,283 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:17,284 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:17,297 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:21,305 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:21,437 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:21,997 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:22,111 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:22,556 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:22,624 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:23,052 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:23,587 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:23,635 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:24,088 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.95 |
| 2026-03-07 06:22:24,088 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:24,098 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:24,251 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:24,252 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:25,873 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.973 |
| 2026-03-07 06:22:25,873 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:25,941 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:26,179 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:26,180 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:27,073 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:22:27,235 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:27,238 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:27,246 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:22:27,349 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:27,639 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer12_mpk.json, llima-compile |
| 2026-03-07 06:22:27,897 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:22:27,920 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv.sima |
| 2026-03-07 06:22:27,946 - 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 06:22:27,946 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv" |
| 2026-03-07 06:22:27,948 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:27,949 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:27,955 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:27,955 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:27,955 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:27,955 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:28,028 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:28,056 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:28,077 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:28,132 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:28,133 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:28,147 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:28,404 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:28,407 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:29,281 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer12_stage1_mla.elf, llima-compile |
| 2026-03-07 06:22:29,676 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14.sima |
| 2026-03-07 06:22:29,694 - 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 06:22:29,694 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14" |
| 2026-03-07 06:22:29,698 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:29,698 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:29,702 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:29,703 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:29,703 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:29,703 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:31,180 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:31,201 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:32,927 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:32,989 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer12_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:22:33,337 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14.sima |
| 2026-03-07 06:22:33,366 - 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 06:22:33,366 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14" |
| 2026-03-07 06:22:33,369 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:33,370 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:33,380 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:33,380 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:33,380 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:33,380 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:33,503 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:33,598 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:33,613 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:34,530 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer11_conv_mpk.json, llima-compile |
| 2026-03-07 06:22:34,903 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14.sima |
| 2026-03-07 06:22:34,923 - 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 06:22:34,923 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14" |
| 2026-03-07 06:22:34,975 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:34,976 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:34,980 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:34,981 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:34,981 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:34,981 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:35,102 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:35,115 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:35,148 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:35,149 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:35,163 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:35,164 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:35,173 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:35,560 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:35,599 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:35,640 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.979 |
| 2026-03-07 06:22:35,640 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:35,699 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:35,700 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:35,715 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:35,718 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:35,778 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:35,778 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:36,496 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:36,515 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:36,887 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:37,573 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:37,645 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:37,688 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:37,795 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:37,870 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:37,871 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:37,880 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:38,308 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:38,322 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:38,331 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:38,338 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:22:38,408 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:38,591 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:38,660 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:38,668 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:38,925 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.943 |
| 2026-03-07 06:22:38,925 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:38,941 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:38,967 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:38,967 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:40,010 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:41,611 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:42,476 - mlc.test_util.test_context - INFO - Compression done in 4s. Compression ratio: 0.971 |
| 2026-03-07 06:22:42,476 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:42,562 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:42,844 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:42,844 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:44,282 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer12_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:22:44,444 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14.sima |
| 2026-03-07 06:22:44,463 - 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 06:22:44,463 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14" |
| 2026-03-07 06:22:44,465 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:44,465 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:44,469 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:44,469 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:44,469 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:44,469 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:44,510 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:44,515 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:44,526 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:44,569 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:44,570 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:44,577 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:45,579 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:45,701 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:45,716 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:45,788 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:46,550 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:46,566 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:46,729 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:22:47,831 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:48,254 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:48,323 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:48,334 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:48,570 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_layer13_conv_mpk.json, llima-compile |
| 2026-03-07 06:22:48,743 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv.sima |
| 2026-03-07 06:22:48,756 - 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 06:22:48,756 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv" |
| 2026-03-07 06:22:48,758 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:48,758 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:48,760 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:48,761 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:48,761 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:48,761 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:50,006 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:50,371 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:50,386 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:50,468 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:50,468 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:50,474 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:50,508 - mlc.test_util.test_context - INFO - Compression done in 2s. Compression ratio: 0.977 |
| 2026-03-07 06:22:50,508 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:50,576 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:50,639 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:50,639 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:53,327 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:22:53,610 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:53,647 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:53,719 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:53,950 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:22:53,951 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:22:53,951 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:22:53,951 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:22:53,951 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:22:53,951 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_pre_layer14_mpk.json, llima-compile |
| 2026-03-07 06:22:54,112 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv.sima |
| 2026-03-07 06:22:54,124 - 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 06:22:54,124 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv" |
| 2026-03-07 06:22:54,126 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:22:54,126 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:22:54,129 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:22:54,129 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:22:54,129 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:22:54,129 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:22:54,174 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:22:54,198 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:22:54,213 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:22:54,228 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:22:54,229 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:22:54,238 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:22:54,394 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:54,794 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:54,864 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:55,134 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:22:55,142 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:22:55,524 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:55,793 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:56,148 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:56,181 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:56,186 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:56,659 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.979 |
| 2026-03-07 06:22:56,659 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:56,675 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:56,696 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:56,696 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:56,765 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:57,228 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:57,383 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:57,402 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:57,418 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:22:57,443 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:57,451 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:22:58,131 - mlc.test_util.test_context - INFO - Compression done in 3s. Compression ratio: 0.969 |
| 2026-03-07 06:22:58,131 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:58,190 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:22:58,199 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:58,296 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.972 |
| 2026-03-07 06:22:58,296 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:58,316 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:58,377 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:58,377 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:58,435 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:58,436 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:22:58,476 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:22:59,018 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:22:59,066 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:22:59,529 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.925 |
| 2026-03-07 06:22:59,530 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:22:59,539 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:22:59,690 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:22:59,691 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:01,764 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:02,274 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:23:03,539 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:03,540 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer14_mpk.json, llima-compile |
| 2026-03-07 06:23:03,731 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0.sima |
| 2026-03-07 06:23:03,743 - 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 06:23:03,743 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0" |
| 2026-03-07 06:23:03,747 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:03,747 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:03,748 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:03,748 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:03,748 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:03,748 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:03,785 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:03,790 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:23:04,334 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:04,335 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer13_conv_stage1_mla.elf, llima-compile |
| 2026-03-07 06:23:04,527 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128.sima |
| 2026-03-07 06:23:04,537 - 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 06:23:04,537 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128" |
| 2026-03-07 06:23:04,541 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:04,541 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:04,542 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:04,542 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:04,542 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:04,542 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:06,693 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:23:06,894 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:06,905 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:07,046 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:07,047 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:07,053 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:07,653 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:23:07,858 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:07,869 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:08,075 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:08,076 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:08,082 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - EV74: 7 |
| 2026-03-07 06:23:08,114 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:08,115 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:08,115 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_pre_layer14_mpk.json, llima-compile |
| 2026-03-07 06:23:08,309 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256.sima |
| 2026-03-07 06:23:08,318 - 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 06:23:08,318 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256" |
| 2026-03-07 06:23:08,321 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:08,321 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:08,322 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:08,323 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:08,323 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:08,323 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:09,079 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 2s. |
| 2026-03-07 06:23:10,554 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:10,560 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:10,981 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_layer15_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:23:11,480 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:23:11,490 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:23:11,559 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384.sima |
| 2026-03-07 06:23:11,569 - 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 06:23:11,569 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384" |
| 2026-03-07 06:23:11,573 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:11,573 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:11,574 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:11,574 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:11,574 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:11,574 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:11,697 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:11,708 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:11,984 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:11,984 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:11,990 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:12,018 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_layer15_conv_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:23:12,213 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512.sima |
| 2026-03-07 06:23:12,222 - 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 06:23:12,223 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512" |
| 2026-03-07 06:23:12,226 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:12,226 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:12,227 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:12,227 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:12,227 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:12,227 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:14,340 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:14,389 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:14,705 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:23:14,913 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:14,924 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:15,258 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:15,259 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:15,264 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:15,301 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:23:15,395 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:15,406 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:15,814 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:15,814 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:15,820 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - EV74: 3 |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:16,386 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_post_layer14_stage1_mla.elf, llima-compile |
| 2026-03-07 06:23:16,578 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640.sima |
| 2026-03-07 06:23:16,589 - 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 06:23:16,590 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640" |
| 2026-03-07 06:23:16,593 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:16,593 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:16,594 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:16,594 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:16,594 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:16,594 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:18,667 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:18,854 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:18,928 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:19,219 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:19,397 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:19,410 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:19,426 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.555 |
| 2026-03-07 06:23:19,427 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:19,427 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 379 |
| 2026-03-07 06:23:19,427 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:19,429 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:19,477 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:19,477 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:20,479 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:24,243 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:24,319 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:24,987 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:25,149 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:25,164 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:25,647 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:25,648 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:25,655 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:25,894 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:26,548 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:26,844 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:26,873 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:26,873 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:26,896 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.391 |
| 2026-03-07 06:23:26,897 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:26,904 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:26,950 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:27,001 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:27,001 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:28,843 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:28,880 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:28,930 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:29,170 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:23:29,461 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:30,564 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token0_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:23:30,757 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768.sima |
| 2026-03-07 06:23:30,767 - 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 06:23:30,767 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768" |
| 2026-03-07 06:23:30,769 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:30,769 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:30,770 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:30,770 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:30,770 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:30,771 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:31,423 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:32,134 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:32,444 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:32,472 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:32,496 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.323 |
| 2026-03-07 06:23:32,496 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:32,503 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:32,535 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token128_mpk.json, llima-compile |
| 2026-03-07 06:23:32,593 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:32,593 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:32,739 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896.sima |
| 2026-03-07 06:23:32,748 - 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 06:23:32,748 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896" |
| 2026-03-07 06:23:32,750 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:32,750 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:32,751 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:32,751 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:32,751 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:32,751 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:33,946 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 376 |
| 2026-03-07 06:23:33,947 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:33,970 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:34,490 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:34,491 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:23:34,632 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:34,943 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:34,971 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:34,998 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.292 |
| 2026-03-07 06:23:34,998 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:35,004 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:35,092 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:35,093 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:36,054 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 418 |
| 2026-03-07 06:23:36,055 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:36,736 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:36,923 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:36,925 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:23:37,447 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:37,667 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token256_stage1_mla.elf, llima-compile |
| 2026-03-07 06:23:37,792 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:37,824 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:37,859 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.269 |
| 2026-03-07 06:23:37,859 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:37,859 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024.sima |
| 2026-03-07 06:23:37,866 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:37,868 - 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 06:23:37,868 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024" |
| 2026-03-07 06:23:37,870 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:37,870 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:37,871 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:37,871 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:37,871 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:37,871 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:37,964 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:37,965 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:38,892 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:39,031 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:39,047 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:39,596 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:39,597 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:39,604 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:39,984 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:39,985 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 2s. |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:40,152 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token384_mpk.json, llima-compile |
| 2026-03-07 06:23:40,344 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152.sima |
| 2026-03-07 06:23:40,353 - 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 06:23:40,353 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152" |
| 2026-03-07 06:23:40,355 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:40,355 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:40,356 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:40,356 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:40,356 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:40,356 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:40,545 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 448 |
| 2026-03-07 06:23:40,545 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:40,825 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:40,919 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:41,223 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:41,335 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:41,351 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:42,329 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:42,330 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:42,338 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:43,476 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 460 |
| 2026-03-07 06:23:43,476 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:43,651 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token512_stage1_mla.elf, llima-compile |
| 2026-03-07 06:23:43,841 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280.sima |
| 2026-03-07 06:23:43,850 - 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 06:23:43,850 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280" |
| 2026-03-07 06:23:43,852 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:43,852 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:43,853 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:43,853 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:43,853 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:43,853 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:46,897 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 432 |
| 2026-03-07 06:23:46,897 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:23:48,187 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:48,307 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:48,328 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:49,010 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:49,011 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:49,020 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:49,326 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:23:50,064 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:23:50,430 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:23:50,462 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:23:50,496 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.258 |
| 2026-03-07 06:23:50,496 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:23:50,503 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:23:50,602 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:23:50,603 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:23:50,860 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:50,965 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:50,988 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:51,734 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:51,734 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:51,744 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:52,715 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:23:52,718 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 2s. |
| 2026-03-07 06:23:54,285 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:23:54,375 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:23:54,398 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:23:54,640 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:54,734 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:55,222 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:23:55,223 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:23:55,232 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:23:56,792 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token640_stage1_mla.elf, llima-compile |
| 2026-03-07 06:23:56,984 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408.sima |
| 2026-03-07 06:23:56,994 - 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 06:23:56,994 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408" |
| 2026-03-07 06:23:56,996 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:23:56,996 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:23:56,997 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:23:56,998 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:23:56,998 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:23:56,998 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:23:58,298 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:23:58,397 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:23:59,328 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 417 |
| 2026-03-07 06:23:59,328 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:24:02,821 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:03,535 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:03,908 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:03,941 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:03,978 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.246 |
| 2026-03-07 06:24:03,978 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:03,984 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:04,091 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:04,091 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:06,267 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:06,268 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 3s. |
| 2026-03-07 06:24:07,214 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:07,952 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:08,123 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:08,277 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:08,332 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:08,365 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:08,407 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.241 |
| 2026-03-07 06:24:08,407 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:08,413 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:08,517 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:08,517 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:08,859 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:24:09,085 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:09,114 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:10,002 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:10,003 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:10,014 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:10,312 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token768_stage1_mla.elf, llima-compile |
| 2026-03-07 06:24:10,409 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:10,504 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536.sima |
| 2026-03-07 06:24:10,513 - 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 06:24:10,513 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536" |
| 2026-03-07 06:24:10,515 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:10,515 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:10,516 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:10,516 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:10,516 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:10,516 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:10,558 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:10,703 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:10,706 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 3s. |
| 2026-03-07 06:24:13,193 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 414 |
| 2026-03-07 06:24:13,193 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:24:14,036 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:14,185 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:14,847 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:14,848 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token896_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:24:15,039 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664.sima |
| 2026-03-07 06:24:15,049 - 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 06:24:15,049 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664" |
| 2026-03-07 06:24:15,050 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:15,050 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:15,051 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:15,051 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:15,051 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:15,051 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:17,342 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 456 |
| 2026-03-07 06:24:17,342 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:24:22,186 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:22,462 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:24:22,660 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:22,688 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:23,055 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:23,640 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:23,641 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:23,652 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:23,657 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:23,706 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:23,751 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.233 |
| 2026-03-07 06:24:23,751 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:23,757 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:23,908 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:23,908 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:24,817 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:25,642 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:26,264 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:26,319 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:26,366 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.23 |
| 2026-03-07 06:24:26,366 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:26,372 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:26,544 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:26,544 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:26,814 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:24:27,003 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:27,034 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:27,051 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:27,053 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:24:28,050 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:28,051 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:28,081 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:28,537 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:29,406 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:29,474 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:29,653 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:30,049 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:30,107 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:30,168 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.225 |
| 2026-03-07 06:24:30,169 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:30,175 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:30,339 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:30,342 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:24:30,353 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:30,353 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:32,975 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:32,976 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1024_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:24:33,175 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792.sima |
| 2026-03-07 06:24:33,185 - 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 06:24:33,185 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792" |
| 2026-03-07 06:24:33,186 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:33,186 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:33,188 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:33,188 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:33,188 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:33,188 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:34,140 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:34,143 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 4s. |
| 2026-03-07 06:24:35,499 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 493 |
| 2026-03-07 06:24:35,499 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:37,376 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1152_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:24:37,570 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920.sima |
| 2026-03-07 06:24:37,579 - 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 06:24:37,579 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920" |
| 2026-03-07 06:24:37,581 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:37,581 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:37,582 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:37,582 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:37,582 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:37,582 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:39,967 - mlc.compiler.model_graph.l1_based - INFO - Trigger large tensor support due to: MLA_0/tuple_einsum_1, 505 |
| 2026-03-07 06:24:39,967 - mlc.compiler.model_graph.l1_based - INFO - Segmenting layers by spatial/channel dimension |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:41,237 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:41,238 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:41,238 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:24:41,238 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:41,238 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:41,238 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1280_stage1_mla.elf, llima-compile |
| 2026-03-07 06:24:41,439 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127.sima |
| 2026-03-07 06:24:41,451 - 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 06:24:41,451 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127" |
| 2026-03-07 06:24:41,453 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:41,453 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:41,454 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:41,454 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:41,454 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:41,454 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:42,016 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:24:42,097 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:42,109 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:42,185 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:42,185 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:42,191 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:44,108 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:44,264 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:44,547 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:24:44,690 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:44,720 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:45,792 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:45,793 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:45,823 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:46,989 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:47,025 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:47,078 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:47,990 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:48,739 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:48,808 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:48,862 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.223 |
| 2026-03-07 06:24:48,862 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:48,868 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:49,085 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:49,086 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:49,268 - mlc.compiler.model_graph.large_tensor_helper - INFO - Setting tile layouts |
| 2026-03-07 06:24:49,353 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:24:49,399 - mlc.compiler.model_graph.large_tensor_helper - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:24:49,431 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:24:49,521 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:24:50,115 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:50,486 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:50,571 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:24:50,572 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:24:50,602 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:24:50,610 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:50,619 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:50,623 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:24:50,623 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:50,623 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:50,654 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:50,654 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:51,248 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:53,599 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:24:53,604 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:24:57,713 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:24:58,147 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:24:58,621 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:24:58,622 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token127_mpk.json, llima-compile |
| 2026-03-07 06:24:58,823 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255.sima |
| 2026-03-07 06:24:58,838 - 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 06:24:58,838 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255" |
| 2026-03-07 06:24:58,840 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:24:58,842 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:24:58,843 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:24:58,843 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:24:58,843 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:24:58,843 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:24:59,059 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:24:59,676 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:24:59,730 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:24:59,788 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.219 |
| 2026-03-07 06:24:59,788 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:24:59,795 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:24:59,960 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:24:59,960 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:24:59,970 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:00,069 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:00,081 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:00,222 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:00,223 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:00,229 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:01,649 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1408_mpk.json, llima-compile |
| 2026-03-07 06:25:01,852 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383.sima |
| 2026-03-07 06:25:01,866 - 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 06:25:01,866 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383" |
| 2026-03-07 06:25:01,867 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:01,867 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:01,868 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:01,869 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:01,869 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:01,869 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:02,670 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:02,749 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:02,761 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:02,966 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:02,967 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:02,973 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:03,402 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:03,407 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 5s. |
| 2026-03-07 06:25:04,227 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:05,132 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:05,864 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:05,927 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:05,989 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.218 |
| 2026-03-07 06:25:05,989 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:05,996 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:06,106 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:06,180 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:06,180 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:06,259 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:07,095 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:07,143 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:09,804 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:09,849 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:10,013 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:10,016 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 6s. |
| 2026-03-07 06:25:10,123 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:10,123 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:10,123 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:10,123 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:10,124 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1536_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:10,327 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511.sima |
| 2026-03-07 06:25:10,336 - 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 06:25:10,336 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511" |
| 2026-03-07 06:25:10,337 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:10,338 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:10,338 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:10,339 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:10,339 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:10,339 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:11,106 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:11,569 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:11,626 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:11,718 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:11,730 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:11,753 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:11,771 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:11,774 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:11,774 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:11,775 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:11,831 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:11,831 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:12,002 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:12,002 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:12,009 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:12,298 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:12,463 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:12,999 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:13,407 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:13,887 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:14,052 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:14,068 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:14,071 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:14,071 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:14,071 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:14,124 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:14,125 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:15,216 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:16,214 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,247 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1664_stage1_mla.elf, llima-compile |
| 2026-03-07 06:25:17,451 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639.sima |
| 2026-03-07 06:25:17,467 - 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 06:25:17,467 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639" |
| 2026-03-07 06:25:17,468 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:17,469 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:17,470 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:17,470 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:17,470 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:17,470 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:17,918 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:17,918 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:17,918 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:17,919 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token255_mpk.json, llima-compile |
| 2026-03-07 06:25:18,125 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767.sima |
| 2026-03-07 06:25:18,136 - 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 06:25:18,136 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767" |
| 2026-03-07 06:25:18,138 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:18,138 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:18,139 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:18,139 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:18,139 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:18,139 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:18,805 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:18,881 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:18,892 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:19,159 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:19,230 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:19,230 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:19,236 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:19,463 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:19,521 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:19,831 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:19,919 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:19,931 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:20,269 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:20,335 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:20,336 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:20,342 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:20,892 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token383_stage1_mla.elf, llima-compile |
| 2026-03-07 06:25:21,154 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895.sima |
| 2026-03-07 06:25:21,163 - 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 06:25:21,163 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895" |
| 2026-03-07 06:25:21,165 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:21,165 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:21,166 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:21,166 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:21,166 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:21,166 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:21,207 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:21,846 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:21,901 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:21,972 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.215 |
| 2026-03-07 06:25:21,973 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:21,979 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:22,144 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:22,144 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:22,878 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:22,950 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:22,962 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:23,432 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:23,432 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:23,438 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:24,083 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:24,603 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:24,814 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:24,832 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:24,836 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:24,836 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:24,836 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:24,897 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:24,897 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:25,590 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:25,593 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 6s. |
| 2026-03-07 06:25:26,150 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:26,172 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:26,200 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:27,221 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:27,755 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:27,811 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:28,186 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:28,922 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:28,985 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:29,063 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.214 |
| 2026-03-07 06:25:29,063 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:29,070 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:29,288 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:29,289 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:30,611 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:30,652 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:30,700 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:31,083 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:31,263 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:31,279 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:31,281 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:31,282 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:31,282 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:31,339 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:31,339 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:32,495 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:32,497 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1792_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:32,702 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023.sima |
| 2026-03-07 06:25:32,710 - 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 06:25:32,710 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023" |
| 2026-03-07 06:25:32,712 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:32,712 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:32,713 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:32,713 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:32,713 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:32,713 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:32,896 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:33,216 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:33,220 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 7s. |
| 2026-03-07 06:25:33,372 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:33,427 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:33,642 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:33,662 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:33,664 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:33,665 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:33,665 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:33,727 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:33,727 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:34,163 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:34,242 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:34,254 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:34,786 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:34,786 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:34,793 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:35,015 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:35,160 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:35,406 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:35,407 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:35,407 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:35,407 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:35,407 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token511_stage1_mla.elf, llima-compile |
| 2026-03-07 06:25:35,623 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151.sima |
| 2026-03-07 06:25:35,632 - 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 06:25:35,632 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151" |
| 2026-03-07 06:25:35,633 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:35,633 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:35,634 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:35,635 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:35,635 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:35,635 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:35,647 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:35,834 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:35,852 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:35,854 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:35,854 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:35,855 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:35,911 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:35,912 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:36,507 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:37,098 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:37,368 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:37,432 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:37,444 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:37,698 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:38,044 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:38,045 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:38,051 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:38,715 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token639_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:38,919 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279.sima |
| 2026-03-07 06:25:38,929 - 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 06:25:38,929 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279" |
| 2026-03-07 06:25:38,930 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:38,930 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:38,932 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:38,932 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:38,932 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:38,932 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:39,945 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token767_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:40,147 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407.sima |
| 2026-03-07 06:25:40,157 - 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 06:25:40,157 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407" |
| 2026-03-07 06:25:40,158 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:40,159 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:40,160 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:40,160 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:40,160 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:40,160 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:40,447 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - EV74: 4 |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:40,578 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n128_cache_token1920_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:40,773 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:40,781 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535.sima |
| 2026-03-07 06:25:40,789 - 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 06:25:40,790 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535" |
| 2026-03-07 06:25:40,791 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:40,791 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:40,792 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:40,792 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:40,792 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:40,792 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:40,850 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:40,862 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:41,524 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:41,525 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:41,531 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:41,899 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:41,955 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:41,964 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:42,023 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:42,035 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:42,266 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:42,337 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:42,349 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:42,509 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token895_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:25:42,713 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663.sima |
| 2026-03-07 06:25:42,722 - 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 06:25:42,722 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663" |
| 2026-03-07 06:25:42,723 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:42,723 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:42,724 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:42,725 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:42,725 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:42,725 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:42,762 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:42,763 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:42,768 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:43,136 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:43,137 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:43,143 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:44,165 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:44,223 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:44,235 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:45,091 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:45,092 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:45,097 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:46,082 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:46,131 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:47,032 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:47,611 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:47,828 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:47,848 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:47,851 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:47,851 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:47,852 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:47,915 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:47,915 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:49,207 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:49,832 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:49,889 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:50,467 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:50,517 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:50,571 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:51,061 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:51,253 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:51,270 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:51,272 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:51,272 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:51,273 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:51,330 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:51,330 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:51,502 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:51,560 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:51,709 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:52,520 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:53,181 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:25:53,230 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:54,082 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1023_stage1_mla.elf, llima-compile |
| 2026-03-07 06:25:54,296 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791.sima |
| 2026-03-07 06:25:54,308 - 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 06:25:54,308 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791" |
| 2026-03-07 06:25:54,310 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:54,310 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:54,311 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:54,311 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:54,311 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:54,311 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:54,925 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:55,001 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:55,211 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:55,492 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:55,685 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:55,703 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:55,705 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:55,705 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:55,705 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:55,765 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:55,765 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:55,767 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:55,989 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:56,009 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:56,012 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:56,012 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:56,012 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:56,076 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:56,076 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:56,184 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:56,256 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:56,268 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:56,796 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:56,964 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:57,186 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:25:57,187 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:25:57,192 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:25:57,195 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1151_stage1_mla.elf, llima-compile |
| 2026-03-07 06:25:57,366 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:57,374 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:57,403 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919.sima |
| 2026-03-07 06:25:57,415 - 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 06:25:57,416 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919" |
| 2026-03-07 06:25:57,417 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:25:57,417 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:25:57,418 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:25:57,418 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:25:57,418 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:25:57,418 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:25:57,592 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:57,612 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:57,615 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:57,615 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:57,616 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:57,680 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:57,680 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:57,824 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:25:58,328 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:25:58,527 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:25:58,544 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:25:58,546 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:25:58,546 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:25:58,546 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:25:58,605 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:25:58,605 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:25:59,035 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:25:59,053 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:25:59,274 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:25:59,332 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:25:59,344 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:25:59,837 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:26:00,282 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:00,340 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:26:00,341 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:26:00,346 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:26:00,709 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:01,566 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1279_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:26:01,770 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047.sima |
| 2026-03-07 06:26:01,779 - 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 06:26:01,779 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047" |
| 2026-03-07 06:26:01,780 - afe.core.compile_networks - INFO - The model is split into 1 segments for MLA and APU |
| 2026-03-07 06:26:01,780 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 1: compiling for MLA |
| 2026-03-07 06:26:01,781 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:26:01,781 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:26:01,781 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:26:01,781 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:26:02,737 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:02,738 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1407_stage1_mla_stats.yaml, llima-compile |
| 2026-03-07 06:26:03,113 - afe.ir.serializer.api - INFO - Loading model from file: CompiledModels/models--LiquidAI--LFM2-VL-450M/sima_files/sdk/models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final.sima |
| 2026-03-07 06:26:03,192 - 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 06:26:03,192 - afe.apis.model - INFO - Compiling quantized net "models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final" |
| 2026-03-07 06:26:03,194 - afe.core.compile_networks - INFO - The model is split into 2 segments for MLA and APU |
| 2026-03-07 06:26:03,194 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 1 of 2: compiling for MLA |
| 2026-03-07 06:26:03,225 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties |
| 2026-03-07 06:26:03,226 - mlc.compiler.model_graph.l1_based - INFO - Checking if layers fit into memory without splitting |
| 2026-03-07 06:26:03,226 - mlc.compiler.model_graph.l1_based - INFO - Generating the list of feasible layouts |
| 2026-03-07 06:26:03,226 - mlc.compiler.model_graph.l1_based - INFO - Setting layer parameters |
| 2026-03-07 06:26:03,271 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:26:03,274 - mlc.compiler.model_graph.l1_based - INFO - Setting tile layouts |
| 2026-03-07 06:26:03,278 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:26:03,292 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:26:03,303 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:03,335 - mlc.compiler.model_graph.l1_based - INFO - Allocating memory for IFM/OFM tensors |
| 2026-03-07 06:26:03,347 - mlc.compiler.model_graph.l1_based - INFO - Get model compilation properties done |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:03,500 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1535_stage1_mla.elf, llima-compile |
| 2026-03-07 06:26:03,517 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:26:03,517 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:26:03,525 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:26:04,403 - afe.backends.mla.afe_to_n2a_compiler.n2a_backend_runner - INFO - Start evaluate process to generate check file |
| 2026-03-07 06:26:04,403 - mlc.kernel.layout - INFO - L2 caching mode: L2CachingMode.NONE |
| 2026-03-07 06:26:04,409 - mlc.compiler.model_graph.l1_based - INFO - Generating model code |
| 2026-03-07 06:26:05,184 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:26:05,245 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:26:05,961 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:05,961 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:05,961 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:05,961 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:05,961 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:05,962 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663_mpk.json, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1663_stage1_mla.elf, llima-compile |
| 2026-03-07 06:26:08,072 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:26:08,122 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:26:10,426 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:26:10,955 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:26:11,179 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:26:11,199 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:26:11,202 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:26:11,202 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:26:11,202 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:26:11,267 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:26:11,267 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:26:11,430 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:26:11,473 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:26:12,580 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:26:12,650 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:26:12,943 - mlc.compiler.model_graph.l1_based - INFO - Generating model code done |
| 2026-03-07 06:26:13,001 - mlc.test_util.test_context - INFO - Scheduling instructions |
| 2026-03-07 06:26:13,141 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:26:13,339 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:26:13,356 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:26:13,359 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:26:13,359 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:26:13,359 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:26:13,418 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:26:13,419 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:26:14,642 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:26:14,997 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:15,012 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:26:16,008 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:26:16,228 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:26:16,269 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:26:17,599 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:17,799 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:17,800 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1791_mpk.json, llima-compile |
| 2026-03-07 06:26:18,392 - mlc.test_util.test_context - INFO - Performing DRAM/L2 synchronization |
| 2026-03-07 06:26:18,984 - mlc.test_util.test_context - INFO - Inserting and merging NOPs |
| 2026-03-07 06:26:19,215 - mlc.test_util.test_context - INFO - Setting IQ sync bits |
| 2026-03-07 06:26:19,235 - mlc.test_util.test_context - INFO - Run compression |
| 2026-03-07 06:26:19,239 - mlc.test_util.test_context - INFO - Compression done in 0s. Compression ratio: 0.404 |
| 2026-03-07 06:26:19,239 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:26:19,239 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:26:19,306 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:26:19,307 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:20,293 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token1919_mpk.json, llima-compile |
| 2026-03-07 06:26:20,671 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:26:20,835 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 0s. |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - EV74: 5 |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - A65 : 0 |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:23,815 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_cache_token2047_mpk.json, llima-compile |
| 2026-03-07 06:26:25,922 - mlc.test_util.test_context - INFO - Compression done in 9s. Compression ratio: 0.918 |
| 2026-03-07 06:26:25,923 - mlc.test_util.test_context - INFO - Re-allocate dram memory |
| 2026-03-07 06:26:26,187 - mlc.test_util.test_context - INFO - Generating metrics |
| 2026-03-07 06:26:26,315 - mlc.test_util.test_context - INFO - Writing report to MLC file |
| 2026-03-07 06:26:26,315 - mlc.test_util.test_context - INFO - Writing instructions to MLC file |
| 2026-03-07 06:26:48,273 - mlc.test_util.test_context - INFO - Code generation done |
| 2026-03-07 06:26:48,287 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Evaluate model and writing chk file done in 1s. |
| 2026-03-07 06:26:49,989 - afe.backends.mla.afe_to_n2a_compiler.n2a_compiler_operations - INFO - Stage 2 of 2: backend is APU |
| 2026-03-07 06:26:49,990 - afe.core.compile_networks - INFO - Stage 2 of 2: compiling for APU |
| 2026-03-07 06:26:56,803 - afe.backends.mpk.interface - INFO - ============================== |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - Compilation summary: |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - Desired batch size: 1 |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - Achieved batch size: 1 |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - Plugin distribution per backend: |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - MLA : 1 |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - EV74: 2 |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - A65 : 1 |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - ------------------------------ |
| 2026-03-07 06:26:56,804 - afe.backends.mpk.interface - INFO - Generated files: models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final_stage1_mla.elf, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final_stage1_mla_stats.yaml, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final_stage2_a65.so, models--LiquidAI--LFM2-VL-450M_language_n1_post_layer15_conv_final_mpk.json, llima-compile |
| 2026-03-07 06:26:59,032 - sima_lmm.model.vision_language_model - INFO - FileGenMode.MODEL_SDK_COMPILE files generation completed. |
| Generated all mode=MODEL_SDK_COMPILE files |
|
|
