Hugging Face's logo

HelloSun
/
Experiment_1

GGUF
conversational
Model card Files
xet
 Community
HelloSun commited on
Commit
db5cbe3
·
verified ·
1 Parent(s): dc0ab8f

Upload verify/all.jsonl with huggingface_hub

Browse files
Files changed (1) hide show
  1. verify/all.jsonl +51 -0
verify/all.jsonl ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"label":"default_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:18:31Z", "avg_ns": 2896052468, "stddev_ns": 74043839, "avg_ts": 176.850179, "stddev_ts": 4.521557, "samples_ns": [ 2948409369, 2843695567 ],"samples_ts": [ 173.653, 180.047 ]}
2
+ {"label":"default_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:18:40Z", "avg_ns": 23253713113, "stddev_ns": 66330097, "avg_ts": 5.504520, "stddev_ts": 0.015701, "samples_ns": [ 23206810900, 23300615327 ],"samples_ts": [ 5.51562, 5.49342 ]}
3
+ {"label":"rec_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:19:27Z", "avg_ns": 2211839224, "stddev_ns": 30772934, "avg_ts": 231.503968, "stddev_ts": 3.220875, "samples_ns": [ 2190079473, 2233598975 ],"samples_ts": [ 233.781, 229.226 ]}
4
+ {"label":"rec_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:19:34Z", "avg_ns": 6538915436, "stddev_ns": 76299999, "avg_ts": 19.576444, "stddev_ts": 0.228430, "samples_ns": [ 6592867683, 6484963189 ],"samples_ts": [ 19.4149, 19.738 ]}
5
+ {"label":"default_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:19:48Z", "avg_ns": 3164518382, "stddev_ns": 71597220, "avg_ts": 161.835393, "stddev_ts": 3.661525, "samples_ns": [ 3215145262, 3113891502 ],"samples_ts": [ 159.246, 164.424 ]}
6
+ {"label":"default_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:19:57Z", "avg_ns": 22984873945, "stddev_ns": 605034650, "avg_ts": 5.570810, "stddev_ts": 0.146641, "samples_ns": [ 22557049868, 23412698023 ],"samples_ts": [ 5.6745, 5.46712 ]}
7
+ {"label":"rec_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:20:43Z", "avg_ns": 2179153036, "stddev_ns": 30980793, "avg_ts": 234.977415, "stddev_ts": 3.340642, "samples_ns": [ 2157246357, 2201059716 ],"samples_ts": [ 237.34, 232.615 ]}
8
+ {"label":"rec_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:20:50Z", "avg_ns": 6384412972, "stddev_ns": 97605748, "avg_ts": 20.051172, "stddev_ts": 0.306545, "samples_ns": [ 6315395332, 6453430613 ],"samples_ts": [ 20.2679, 19.8344 ]}
9
+ {"label":"default_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:21:03Z", "avg_ns": 2801677647, "stddev_ns": 24052050, "avg_ts": 182.754382, "stddev_ts": 1.568916, "samples_ns": [ 2784670362, 2818684933 ],"samples_ts": [ 183.864, 181.645 ]}
10
+ {"label":"default_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 96, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": -1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:21:12Z", "avg_ns": 22825873420, "stddev_ns": 196256427, "avg_ts": 5.607879, "stddev_ts": 0.048216, "samples_ns": [ 22687099169, 22964647671 ],"samples_ts": [ 5.64197, 5.57378 ]}
11
+ {"label":"rec_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:21:58Z", "avg_ns": 2148734965, "stddev_ns": 23947856, "avg_ts": 238.294535, "stddev_ts": 2.655805, "samples_ns": [ 2165668594, 2131801337 ],"samples_ts": [ 236.417, 240.172 ]}
12
+ {"label":"rec_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:22:05Z", "avg_ns": 6357312700, "stddev_ns": 26145812, "avg_ts": 20.134464, "stddev_ts": 0.082807, "samples_ns": [ 6375800581, 6338824819 ],"samples_ts": [ 20.0759, 20.193 ]}
13
+ {"label":"rp1_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:22:18Z", "avg_ns": 3022829190, "stddev_ns": 1168373660, "avg_ts": 183.051224, "stddev_ts": 70.752337, "samples_ns": [ 3848994127, 2196664254 ],"samples_ts": [ 133.022, 233.081 ]}
14
+ {"label":"rp1_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:22:26Z", "avg_ns": 6858373804, "stddev_ns": 122852166, "avg_ts": 18.666311, "stddev_ts": 0.334364, "samples_ns": [ 6771504244, 6945243365 ],"samples_ts": [ 18.9027, 18.4299 ]}
15
+ {"label":"rp0_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:22:41Z", "avg_ns": 2151018750, "stddev_ns": 39451011, "avg_ts": 238.066789, "stddev_ts": 4.366292, "samples_ns": [ 2123122672, 2178914828 ],"samples_ts": [ 241.154, 234.979 ]}
16
+ {"label":"rp0_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:22:47Z", "avg_ns": 6707200771, "stddev_ns": 285957096, "avg_ts": 19.101327, "stddev_ts": 0.814373, "samples_ns": [ 6909402973, 6504998569 ],"samples_ts": [ 18.5255, 19.6772 ]}
17
+ {"label":"fa0_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:23:01Z", "avg_ns": 2434905625, "stddev_ns": 41095773, "avg_ts": 210.305044, "stddev_ts": 3.549475, "samples_ns": [ 2405846567, 2463964684 ],"samples_ts": [ 212.815, 207.795 ]}
18
+ {"label":"fa0_r1","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:23:09Z", "avg_ns": 6473755586, "stddev_ns": 3027456, "avg_ts": 19.772142, "stddev_ts": 0.009246, "samples_ns": [ 6471614851, 6475896321 ],"samples_ts": [ 19.7787, 19.7656 ]}
19
+ {"label":"rp1_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:23:22Z", "avg_ns": 3042937128, "stddev_ns": 1015165990, "avg_ts": 178.173694, "stddev_ts": 59.441213, "samples_ns": [ 3760767884, 2325106372 ],"samples_ts": [ 136.142, 220.205 ]}
20
+ {"label":"rp1_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:23:30Z", "avg_ns": 6694142877, "stddev_ns": 90237359, "avg_ts": 19.122931, "stddev_ts": 0.257778, "samples_ns": [ 6757950326, 6630335428 ],"samples_ts": [ 18.9407, 19.3052 ]}
21
+ {"label":"rp0_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:23:45Z", "avg_ns": 2183818497, "stddev_ns": 1947262, "avg_ts": 234.451812, "stddev_ts": 0.208935, "samples_ns": [ 2182442368, 2185194627 ],"samples_ts": [ 234.6, 234.304 ]}
22
+ {"label":"rp0_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:23:51Z", "avg_ns": 6507882018, "stddev_ns": 237435143, "avg_ts": 19.681556, "stddev_ts": 0.718067, "samples_ns": [ 6339990038, 6675773999 ],"samples_ts": [ 20.1893, 19.1738 ]}
23
+ {"label":"fa0_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:24:05Z", "avg_ns": 2450355492, "stddev_ns": 30181499, "avg_ts": 208.965125, "stddev_ts": 2.573857, "samples_ns": [ 2471696978, 2429014007 ],"samples_ts": [ 207.145, 210.785 ]}
24
+ {"label":"fa0_r2","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:24:12Z", "avg_ns": 6609799266, "stddev_ns": 109382097, "avg_ts": 19.367839, "stddev_ts": 0.320508, "samples_ns": [ 6532454443, 6687144089 ],"samples_ts": [ 19.5945, 19.1412 ]}
25
+ {"label":"rp1_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:24:26Z", "avg_ns": 2996340737, "stddev_ns": 1145926135, "avg_ts": 184.357306, "stddev_ts": 70.505952, "samples_ns": [ 3806632877, 2186048598 ],"samples_ts": [ 134.502, 234.213 ]}
26
+ {"label":"rp1_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": true, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:24:34Z", "avg_ns": 6744905133, "stddev_ns": 146218080, "avg_ts": 18.981747, "stddev_ts": 0.411492, "samples_ns": [ 6848296929, 6641513337 ],"samples_ts": [ 18.6908, 19.2727 ]}
27
+ {"label":"rp0_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:24:48Z", "avg_ns": 2141245673, "stddev_ns": 21286514, "avg_ts": 239.124967, "stddev_ts": 2.377185, "samples_ns": [ 2156297512, 2126193834 ],"samples_ts": [ 237.444, 240.806 ]}
28
+ {"label":"rp0_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:24:55Z", "avg_ns": 6532004534, "stddev_ns": 222490301, "avg_ts": 19.607196, "stddev_ts": 0.667852, "samples_ns": [ 6374680154, 6689328915 ],"samples_ts": [ 20.0794, 19.135 ]}
29
+ {"label":"fa0_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:25:08Z", "avg_ns": 2554917598, "stddev_ns": 17058342, "avg_ts": 200.402319, "stddev_ts": 1.338020, "samples_ns": [ 2566979668, 2542855528 ],"samples_ts": [ 199.456, 201.348 ]}
30
+ {"label":"fa0_r3","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 0, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:25:16Z", "avg_ns": 6623366996, "stddev_ns": 215826530, "avg_ts": 19.335784, "stddev_ts": 0.630069, "samples_ns": [ 6775979399, 6470754593 ],"samples_ts": [ 18.8903, 19.7813 ]}
31
+ {"label":"thr_4","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 4, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:25:30Z", "avg_ns": 7459368624, "stddev_ns": 93449109, "avg_ts": 68.645651, "stddev_ts": 0.854232, "samples_ns": [ 7394225774, 7566438609, 7417441490 ],"samples_ts": [ 69.2432, 67.6672, 69.0265 ]}
32
+ {"label":"thr_4","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 4, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:26:00Z", "avg_ns": 21406688103, "stddev_ns": 219963764, "avg_ts": 5.979858, "stddev_ts": 0.061105, "samples_ns": [ 21255308356, 21659005110, 21305750845 ],"samples_ts": [ 6.02203, 5.90978, 6.00777 ]}
33
+ {"label":"thr_8","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 8, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:27:04Z", "avg_ns": 3870019505, "stddev_ns": 117976948, "avg_ts": 132.381302, "stddev_ts": 4.046578, "samples_ns": [ 3984976688, 3749238505, 3875843322 ],"samples_ts": [ 128.483, 136.561, 132.1 ]}
34
+ {"label":"thr_8","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 8, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:27:20Z", "avg_ns": 11308474708, "stddev_ns": 140450612, "avg_ts": 11.320114, "stddev_ts": 0.141279, "samples_ns": [ 11425148200, 11347692175, 11152583751 ],"samples_ts": [ 11.2034, 11.2798, 11.4772 ]}
35
+ {"label":"thr_12","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 12, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:27:54Z", "avg_ns": 2829962854, "stddev_ns": 29473475, "avg_ts": 180.934194, "stddev_ts": 1.885548, "samples_ns": [ 2858892016, 2799973756, 2831022791 ],"samples_ts": [ 179.09, 182.859, 180.853 ]}
36
+ {"label":"thr_12","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 12, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:28:06Z", "avg_ns": 8057590702, "stddev_ns": 7961011, "avg_ts": 15.885652, "stddev_ts": 0.015701, "samples_ns": [ 8048702015, 8064063116, 8060006976 ],"samples_ts": [ 15.9032, 15.8729, 15.8809 ]}
37
+ {"label":"thr_16","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:28:30Z", "avg_ns": 2161168381, "stddev_ns": 9315613, "avg_ts": 236.911818, "stddev_ts": 1.019174, "samples_ns": [ 2171671035, 2153904795, 2157929314 ],"samples_ts": [ 235.763, 237.708, 237.264 ]}
38
+ {"label":"thr_16","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:28:39Z", "avg_ns": 6615667386, "stddev_ns": 374008321, "avg_ts": 19.388119, "stddev_ts": 1.064280, "samples_ns": [ 6450666116, 7043802443, 6352533600 ],"samples_ts": [ 19.8429, 18.172, 20.1494 ]}
39
+ {"label":"thr_20","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 20, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:28:59Z", "avg_ns": 2385875654, "stddev_ns": 12238922, "avg_ts": 214.600039, "stddev_ts": 1.104022, "samples_ns": [ 2393789981, 2371779013, 2392057970 ],"samples_ts": [ 213.887, 215.872, 214.042 ]}
40
+ {"label":"thr_20","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 20, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:29:09Z", "avg_ns": 11521812941, "stddev_ns": 86350071, "avg_ts": 11.109781, "stddev_ts": 0.083623, "samples_ns": [ 11573340188, 11569975240, 11422123395 ],"samples_ts": [ 11.0599, 11.0631, 11.2063 ]}
41
+ {"label":"thr_24","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 24, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:29:44Z", "avg_ns": 2421976851, "stddev_ns": 21942765, "avg_ts": 211.409078, "stddev_ts": 1.906737, "samples_ns": [ 2446921002, 2413354730, 2405654823 ],"samples_ts": [ 209.243, 212.153, 212.832 ]}
42
+ {"label":"thr_24","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 24, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:29:54Z", "avg_ns": 12138215180, "stddev_ns": 164464279, "avg_ts": 10.546508, "stddev_ts": 0.143979, "samples_ns": [ 12217095634, 11949169122, 12248380784 ],"samples_ts": [ 10.4771, 10.712, 10.4504 ]}
43
+ {"label":"kv_f16","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:30:30Z", "avg_ns": 2161071726, "stddev_ns": 10708451, "avg_ts": 236.923352, "stddev_ts": 1.170622, "samples_ns": [ 2154893144, 2173436548, 2154885488 ],"samples_ts": [ 237.599, 235.572, 237.6 ]}
44
+ {"label":"kv_f16","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:30:39Z", "avg_ns": 6370545022, "stddev_ns": 102343401, "avg_ts": 20.095900, "stddev_ts": 0.320063, "samples_ns": [ 6323926551, 6487897920, 6299810595 ],"samples_ts": [ 20.2406, 19.729, 20.3181 ]}
45
+ {"label":"kv_q8_0","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "q8_0", "type_v": "q8_0", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:30:59Z", "avg_ns": 2628032292, "stddev_ns": 64396870, "avg_ts": 194.900038, "stddev_ts": 4.742663, "samples_ns": [ 2614845484, 2571249576, 2698001818 ],"samples_ts": [ 195.805, 199.125, 189.77 ]}
46
+ {"label":"kv_q8_0","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "q8_0", "type_v": "q8_0", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:31:09Z", "avg_ns": 6508135301, "stddev_ns": 167318189, "avg_ts": 19.676477, "stddev_ts": 0.512580, "samples_ns": [ 6575654397, 6631143891, 6317607616 ],"samples_ts": [ 19.4657, 19.3029, 20.2608 ]}
47
+ {"label":"kv_q4_0","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "q4_0", "type_v": "q4_0", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 512, "n_gen": 0, "n_depth": 0, "test_time": "2026-10-27T03:31:29Z", "avg_ns": 2663669033, "stddev_ns": 22484208, "avg_ts": 192.225165, "stddev_ts": 1.615765, "samples_ns": [ 2647005933, 2654758325, 2689242841 ],"samples_ts": [ 193.426, 192.861, 190.388 ]}
48
+ {"label":"kv_q4_0","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "q4_0", "type_v": "q4_0", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 128, "n_depth": 0, "test_time": "2026-10-27T03:31:40Z", "avg_ns": 6498354518, "stddev_ns": 120122864, "avg_ts": 19.701789, "stddev_ts": 0.364729, "samples_ns": [ 6375490482, 6615534397, 6504038675 ],"samples_ts": [ 20.0769, 19.3484, 19.6801 ]}
49
+ {"label":"depth_2048","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 64, "n_depth": 2048, "test_time": "2026-10-27T03:32:00Z", "avg_ns": 3741686104, "stddev_ns": 109400547, "avg_ts": 17.114188, "stddev_ts": 0.492490, "samples_ns": [ 3667611353, 3690105342, 3867341618 ],"samples_ts": [ 17.45, 17.3437, 16.5488 ]}
50
+ {"label":"depth_4096","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 64, "n_depth": 4096, "test_time": "2026-10-27T03:32:23Z", "avg_ns": 4252776069, "stddev_ns": 76351181, "avg_ts": 15.052204, "stddev_ts": 0.268259, "samples_ns": [ 4229488018, 4338059410, 4190780780 ],"samples_ts": [ 15.1319, 14.7531, 15.2716 ]}
51
+ {"label":"depth_8192","build_commit": "1268edac5", "build_number": 1, "cpu_info": "Intel(R) Xeon(R) Platinum 8559C", "gpu_info": "", "backends": "CPU", "model_filename": "./openpangu-vl-artifacts/openPangu-VL-7B-Q4_K_M.gguf", "model_type": "pangu_vl 7B Q4_K - Medium", "model_size": 4929676288, "model_n_params": 8030887936, "n_batch": 2048, "n_ubatch": 512, "n_threads": 16, "cpu_mask": "0x0", "cpu_strict": false, "poll": 50, "type_k": "f16", "type_v": "f16", "n_gpu_layers": -1, "n_cpu_moe": 0, "split_mode": "layer", "main_gpu": 0, "no_kv_offload": false, "flash_attn": 1, "devices": "auto", "tensor_split": "0.00", "tensor_buft_overrides": "none", "load_mode": "auto", "lazy_mode": "auto", "embeddings": false, "no_op_offload": 0, "no_host": false, "repack": false, "fit_target": 0, "fit_min_ctx": 0, "n_prompt": 0, "n_gen": 64, "n_depth": 8192, "test_time": "2026-10-27T03:33:00Z", "avg_ns": 5651508434, "stddev_ns": 80739764, "avg_ts": 11.325952, "stddev_ts": 0.161845, "samples_ns": [ 5570524537, 5651999077, 5732001690 ],"samples_ts": [ 11.489, 11.3234, 11.1654 ]}