diff --git "a/perf-df-awq-1xA10.csv" "b/perf-df-awq-1xA10.csv" --- "a/perf-df-awq-1xA10.csv" +++ "b/perf-df-awq-1xA10.csv" @@ -621,7 +621,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -639,13 +639,13 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3620, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3704, in from_pretrained config = cls._autoset_attn_implementation( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1478, in _autoset_attn_implementation + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1490, in _autoset_attn_implementation config = cls._check_and_enable_sdpa( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1644, in _check_and_enable_sdpa + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1656, in _check_and_enable_sdpa raise ValueError( ValueError: DeciLMForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` @@ -2762,7 +2762,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: OPTForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -2782,33 +2782,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 615, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 614, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -4319,7 +4319,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: GPTNeoXForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -4339,33 +4339,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 615, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 614, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -4568,7 +4568,7 @@ AssertionError ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-awq-gemm-eager,pytorch,2.3.0+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,EleutherAI/polyglot-ko-12.8b,EleutherAI/polyglot-ko-12.8b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.215-203.850.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.40.2,,0.30.1,,,,1.19.2,,,,0.11.0,,,,,,MB,1740.263424,9941.024768,0.0,9294.577664,8910.102528,s,10,10.642888305664064,1.0642888305664062,0.0008435144191809837,1.06450927734375,1.0651936889648437,1.065327557373047,1.0654346520996094,"[1.065136474609375, 1.06546142578125, 1.063631591796875, 1.063218505859375, 1.06368701171875, 1.0628612060546876, 1.0644658203125, 1.064552734375, 1.0647095947265626, 1.0651639404296875]",tokens/s,240.5362084498799,kWh,1.2557922783825134e-05,6.880571250694628e-06,6.097363211219897e-05,8.041212614671872e-05,tokens/kWh,3183599.4428614704,MB,1740.86144,9941.024768,0.0,9294.577664,9220.867072,s,10,630.7354296875,63.07354296875,0.00490218127220892,63.07426171875,63.081011718750005,63.081064453125,63.081106640625,"[63.0750703125, 63.07307421875, 63.081, 63.075046875, 63.07416015625, 63.0811171875, 63.07436328125, 63.0665078125, 63.06765234375, 63.0674375]",tokens/s,0.9988340124037992,kWh,0.0007445980064074199,0.0004081038876869388,0.003603297882636,0.004755999776730358,tokens/kWh,13246.426189555263,,s,629,639.312124511719,1.016394474581429,0.12643667238363906,1.0011002807617186,1.0017415283203124,1.002017578125,2.065029130859375,"[1.0008678588867188, 1.0004398193359374, 1.001291748046875, 1.0011023559570313, 1.0010675048828126, 1.0010203857421875, 1.0008780517578124, 1.000616943359375, 1.0006917114257812, 1.0008555297851562, 1.0007982177734376, 1.000685546875, 1.0008023071289063, 1.000985595703125, 1.0008780517578124, 1.0010081787109375, 1.0006251220703124, 1.0007255249023437, 1.000806396484375, 1.0009149169921876, 1.0006405029296874, 1.0007285766601564, 1.0012835693359374, 1.0011812133789062, 1.0010265502929687, 1.0012303466796875, 1.0011586303710938, 1.0011637573242187, 1.0010664672851564, 1.0014730224609374, 1.0011238403320313, 1.0014033813476562, 1.0011033325195313, 1.0013900756835938, 1.0010501098632814, 1.0017617797851563, 1.0014638061523438, 1.0016256103515624, 1.0009415893554687, 1.0013726806640626, 1.000985595703125, 1.0008145751953126, 1.001359375, 1.0016143188476563, 1.0011217651367188, 1.0013255615234375, 1.0009302978515624, 1.001133056640625, 1.0009508056640626, 1.001533447265625, 1.0019143676757813, 1.0015897827148437, 1.0020361938476563, 1.0020464477539062, 1.00206591796875, 1.0012640991210937, 1.0014044189453124, 1.001734130859375, 1.0015846557617187, 1.00234033203125, 1.0014207763671874, 1.0013255615234375, 2.067092529296875, 1.0008361206054688, 1.0011576538085938, 1.0006384887695312, 1.001079833984375, 1.0007879638671875, 1.0012293090820312, 1.0010501098632814, 1.001322509765625, 1.0013265991210938, 1.0013900756835938, 1.000838134765625, 1.001486328125, 1.0010675048828126, 1.0011484375, 1.0016091918945313, 1.001449462890625, 1.0011023559570313, 1.0005330200195313, 1.0008074340820312, 1.0010623779296874, 1.0006220703125, 1.0009927978515625, 1.0007716064453125, 1.001026611328125, 1.0008770141601562, 1.0014136352539063, 1.0009508056640626, 1.0009508056640626, 1.00097021484375, 1.001290771484375, 1.00075830078125, 1.0016583862304687, 1.0007131958007813, 1.0006619873046876, 1.0005339965820312, 1.0010706176757813, 1.0007859497070313, 1.0006456298828126, 1.0009036865234375, 1.00139111328125, 1.0010776977539062, 1.0015672607421875, 1.0008135375976563, 1.0011463623046875, 1.000784912109375, 1.0035189819335937, 1.00113818359375, 1.0015252685546876, 1.0012507934570312, 1.0016573486328124, 1.0014893798828124, 1.001533447265625, 1.0010839233398436, 1.0011525268554688, 1.0011566162109375, 1.0015396118164062, 1.0014955444335938, 1.0011074829101563, 1.0009037475585938, 1.0016132202148438, 1.0009343872070313, 1.002029052734375, 2.06504150390625, 1.0005913696289062, 1.0010398559570313, 1.00097021484375, 1.0005196533203125, 1.0015057983398437, 1.0011084594726563, 1.0007337036132813, 1.0012190551757814, 1.0008473510742188, 1.0012252197265625, 1.0008402099609375, 1.0011617431640625, 1.0010194091796876, 1.0008719482421875, 1.0009927978515625, 1.0008237915039062, 1.00075927734375, 1.0011688842773439, 1.000806396484375, 1.0008606567382812, 1.0010132446289062, 1.0011740112304688, 1.00221337890625, 1.0005995483398438, 1.000721435546875, 1.00090673828125, 1.0007039794921875, 1.001343994140625, 1.001470947265625, 1.0014852905273437, 1.00143408203125, 1.0017413330078124, 1.002018798828125, 1.0018693237304688, 1.0017587280273437, 1.0011954956054687, 1.0007203979492187, 1.0011064453125, 1.00105419921875, 1.0013839111328124, 1.0012108764648437, 1.001881591796875, 1.0012119140625, 1.0016245727539062, 1.0019154052734376, 1.0013501586914062, 1.0026547241210937, 1.0016737060546874, 1.0011678466796874, 1.00153857421875, 1.0014085083007813, 1.0018191528320313, 1.001275390625, 1.0014157104492187, 1.0018908081054687, 1.0011791381835937, 1.0009927978515625, 1.0019522705078125, 1.0013368530273437, 1.0016307373046875, 1.0016552734375, 1.001829345703125, 2.064997314453125, 1.001175048828125, 1.0011033935546876, 1.0011688232421876, 1.001164794921875, 1.0007736206054687, 1.001064453125, 1.0007973022460936, 1.00076025390625, 1.0006077270507812, 1.0007183227539063, 1.0011361083984376, 1.0011351318359376, 1.0006005859375, 1.0008688354492188, 1.000642578125, 1.00075830078125, 1.0010562744140625, 1.001421875, 1.0009476318359376, 1.000995849609375, 1.0008811645507814, 1.001438232421875, 1.000816650390625, 1.0011361083984376, 1.0006569213867187, 1.0008370971679688, 1.0008350830078125, 1.00099072265625, 1.0007787475585936, 1.0014464111328125, 1.0013634643554687, 1.0017720336914062, 1.001069580078125, 1.001523193359375, 1.0014351196289062, 1.002029052734375, 1.001865234375, 1.0020095825195312, 1.0007572631835937, 1.0010081176757812, 1.0010040283203125, 1.000943603515625, 1.001080810546875, 1.0031155395507811, 1.000857666015625, 1.00139208984375, 1.001006103515625, 1.0015139770507813, 1.0011719970703126, 1.0016010131835937, 1.0014443359375, 1.0011443481445312, 1.0013952026367188, 1.0011340942382811, 1.0010603637695312, 1.0010203857421875, 1.0011033935546876, 1.0015077514648438, 1.00125390625, 1.002071044921875, 1.0015109252929688, 1.0014412841796876, 2.065207275390625, 1.0004561767578124, 1.0008197021484375, 1.0008104858398437, 1.0015027465820312, 1.0008872680664063, 1.0009497680664063, 1.0020311279296874, 1.0013726806640626, 1.0015405883789064, 1.00103271484375, 1.0013511962890624, 1.0011668701171874, 1.0011986083984374, 1.00210791015625, 1.0007920532226562, 1.0011351318359376, 1.0023803100585937, 1.0009579467773437, 1.0010460205078124, 1.0012252197265625, 1.0007500610351563, 1.001091064453125, 1.0007879638671875, 1.0017576904296874, 1.0010828857421874, 1.0014474487304688, 1.0011094970703125, 1.0007890014648437, 1.0009169921875, 1.0007787475585936, 1.0004838256835937, 1.0008545532226563, 1.0006609497070313, 1.0009682006835938, 1.0007777099609374, 1.0011300048828125, 1.00124365234375, 1.000796142578125, 1.0009343872070313, 1.0009251708984375, 1.000722412109375, 1.0007971801757813, 1.0006896362304687, 1.0014924926757813, 1.0007080688476562, 1.0007705688476562, 1.0013204345703124, 1.001218017578125, 1.0010511474609376, 1.0009517822265626, 1.0008607177734374, 1.0009199829101563, 1.00119140625, 1.0017146606445313, 1.0010194091796876, 1.0013358154296874, 1.003936767578125, 1.001802734375, 1.0014443359375, 1.0012477416992187, 1.0015396118164062, 1.0018938598632812, 2.064691162109375, 1.0003916625976563, 1.0009784545898437, 1.0010337524414064, 1.0008944702148437, 1.00107470703125, 1.0008616943359374, 1.001354248046875, 1.00143408203125, 1.0012088623046875, 1.0020802612304687, 1.0009651489257811, 1.001312255859375, 1.0014505004882812, 1.0007481079101563, 1.0010009155273438, 1.0012723388671876, 1.0009712524414063, 1.0011033325195313, 1.0011371459960938, 1.0015723266601562, 1.0016163940429688, 1.0008381958007813, 1.0017801513671876, 1.0015897827148437, 1.0014259033203126, 1.0019799194335937, 1.0012518310546874, 1.0014515380859375, 1.0014484252929687, 1.001759765625, 1.00107568359375, 1.0014505004882812, 1.0014423217773438, 1.001112548828125, 1.0011678466796874, 1.0010286254882812, 1.0009210815429688, 1.0033643798828125, 1.0010767211914062, 1.001469970703125, 1.0009139404296874, 1.001143310546875, 1.00132763671875, 1.0011791381835937, 1.001169921875, 1.0012620849609375, 1.0009456787109374, 1.0010347290039063, 1.00092724609375, 1.0015047607421874, 1.00096923828125, 1.0009251708984375, 1.0013870239257812, 1.0013132934570312, 1.0014105834960938, 1.0011658325195312, 1.0012948608398438, 1.0014935302734376, 1.0009978637695311, 1.0015344848632812, 1.001290771484375, 1.0020157470703126, 2.0664228515625, 1.001302001953125, 1.000827880859375, 1.0009938354492187, 1.0006937255859376, 1.0006784057617188, 1.000911865234375, 1.0006487426757813, 1.0011852416992189, 1.0008043823242188, 1.0012477416992187, 1.0013726806640626, 1.0013101806640625, 1.0008023071289063, 1.0008514404296875, 1.0008248291015625, 1.0011658325195312, 1.001006103515625, 1.0012498168945312, 1.0011207885742188, 1.0012610473632813, 1.0012415771484375, 1.0010951538085937, 1.0023833618164062, 1.0010685424804688, 1.0005288696289063, 1.000853515625, 1.000543212890625, 1.0008811645507814, 1.0009784545898437, 1.0013153076171875, 1.0010859375, 1.0009425659179687, 1.0009343872070313, 1.000748046875, 1.000453125, 1.0013388671875, 1.00075927734375, 1.000826904296875, 1.00143310546875, 1.0011207885742188, 1.0015027465820312, 1.001175048828125, 1.0009395141601563, 1.0011279296875, 1.0010224609375, 1.0013798217773437, 1.0013460693359375, 1.00193896484375, 1.00166552734375, 1.0014197998046874, 1.0018252563476562, 1.0011893920898438, 1.0011268920898437, 1.0015764770507813, 1.0016574096679687, 1.0018969116210938, 1.0011760864257813, 1.0016829223632813, 1.0013952026367188, 1.0015078125, 1.0017791748046876, 1.001365478515625, 2.065933349609375, 1.00049609375, 1.0008944702148437, 1.00080126953125, 1.0008135375976563, 1.0008340454101563, 1.0008822021484376, 1.001302001953125, 1.0008237915039062, 1.0006292724609376, 1.0011105346679687, 1.0010654907226562, 1.0013255615234375, 1.0006692504882813, 1.0007797241210938, 1.0010091552734375, 1.0015211791992187, 1.0007367553710937, 1.0007234497070312, 1.0005473022460938, 1.0006712036132812, 1.00077978515625, 1.0011351318359376, 1.001279541015625, 1.0010480346679687, 1.0010921020507813, 1.000892333984375, 1.001080810546875, 1.00067431640625, 1.0010296020507812, 1.0010767822265625, 1.0005595703125, 1.0013173828125, 1.0010501098632814, 1.0013409423828126, 1.0012006225585937, 1.0011443481445312, 1.0012139282226562, 1.0008053588867187, 1.0007408447265624, 1.0009384765625, 1.0005883178710937, 1.0010715942382813, 1.0009569091796875, 1.0010900268554688, 1.0009027099609376, 1.000849365234375, 1.0009651489257811, 1.0008524780273438, 1.001027587890625, 1.0014033813476562, 1.0012498168945312, 1.00135009765625, 1.001059326171875, 1.0014095458984376, 1.0020269775390624, 1.0020638427734374, 1.0020095825195312, 1.0013235473632813, 1.0012037353515626, 1.0014484252929687, 1.0009476928710936, 1.0012733154296876, 2.066282470703125, 1.0010859375, 1.0004767456054688, 1.0010264892578125, 1.0008135375976563, 1.0007572631835937, 1.0011146240234374, 1.0009620361328124, 1.0013931274414063, 1.0009630737304687, 1.0009375, 1.0007900390625, 1.0008340454101563, 1.0009886474609375, 1.0007900390625, 1.0007756958007812, 1.0007982177734376, 1.0019952392578124, 1.0009794311523437, 1.0006138916015626, 1.0008309936523438, 1.0008893432617187, 1.0010368041992188, 1.0011248779296875, 1.0005238037109374, 1.0006343383789063, 1.0008145751953126, 1.0007039794921875, 1.0007521362304688, 1.0009896850585938, 1.0010880126953126, 1.0012303466796875, 1.00091796875, 1.0013450317382813, 1.000784912109375, 1.0007367553710937, 1.0009825439453126, 1.0009108276367187, 1.001238525390625, 1.0008944702148437, 1.0009989013671876, 1.0009886474609375, 1.0009651489257811, 1.0011422729492188, 1.0012374877929688, 1.0012569580078126, 1.00099072265625, 1.0008780517578124, 1.0009682006835938, 1.0007080688476562, 1.001248779296875, 1.0015518798828125, 1.0010582885742187, 1.0014033813476562, 1.001016357421875, 1.0009476318359376, 1.0012354736328124, 1.0020833129882813, 1.0019143676757813, 1.0018539428710938, 1.0019850463867188, 1.0015999755859375, 1.0015641479492188, 2.0665712890625, 1.0007060546875, 1.0006599731445311, 1.00056884765625, 1.000685546875, 1.0005678100585937, 1.0009722900390625, 1.000806396484375, 1.0007203979492187, 1.0008207397460938, 1.0005872802734375, 1.0005545043945312, 1.00069384765625, 1.0008534545898438, 1.0013716430664064, 1.0008811645507814, 1.0013511962890624, 1.0014893798828124, 1.0010726318359375, 1.0017157592773438, 1.0011299438476562, 1.00086376953125, 1.00071728515625, 1.0005729370117187, 1.0008995971679688, 1.0005995483398438, 1.0024898681640626, 1.0008678588867188, 1.000722412109375, 1.0012149658203124, 1.0008237915039062, 1.00071728515625, 1.00101220703125, 1.0010562744140625, 1.0010572509765625, 1.00170751953125, 1.00140234375, 1.000974365234375, 1.0007828369140626, 1.0009139404296874, 1.0011300048828125, 1.0010675048828126, 1.0013562622070313, 1.000953857421875, 1.0009722900390625, 1.0011146240234374, 1.0015631103515625, 1.000975341796875, 1.0017495727539063, 1.0013224487304688, 1.0012518310546874, 1.001439208984375, 1.0012415771484375, 1.0010685424804688, 1.0010286254882812, 1.00105419921875, 1.0016696166992187, 1.0011002807617186, 1.0011443481445312, 1.0017423095703124, 1.0011033325195313, 1.0011924438476563, 1.0012406005859376]",tokens/s,0.9838699688049951,,,,, -4bit-awq-gemm-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemm-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -4588,27 +4588,27 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -7296,7 +7296,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -7304,7 +7304,7 @@ NameError: name 'exlv2_ext' is not defined File ""/workspace/optimum_benchmark/launchers/process/launcher.py"", line 66, in launch raise ChildProcessError(response[""traceback""]) ChildProcessError: Traceback (most recent call last): - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 612, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 613, in resolve_trust_remote_code answer = input( EOFError: EOF when reading a line @@ -7321,9 +7321,9 @@ Traceback (most recent call last): self.create_no_weights_model() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 210, in create_no_weights_model meta_model = self.automodel_class.from_config(self.pretrained_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 418, in from_config + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 419, in from_config trust_remote_code = resolve_trust_remote_code( - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 625, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 626, in resolve_trust_remote_code raise ValueError( ValueError: The repository for Deci/DeciLM-7B contains custom code which must be executed to correctly load the model. You can inspect the repository content at https://hf.co/Deci/DeciLM-7B. Please pass the argument `trust_remote_code=True` to allow custom code to be run. @@ -9237,7 +9237,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: OPTForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -9255,15 +9255,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device]) @@ -10610,7 +10610,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: GPTNeoXForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -10628,15 +10628,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device]) @@ -13813,7 +13813,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -13833,9 +13833,9 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -16236,7 +16236,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -16256,33 +16256,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 416, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 414, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -18160,7 +18160,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -18180,33 +18180,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 416, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 414, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -18583,7 +18583,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -18603,27 +18603,27 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -19090,7 +19090,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -19110,9 +19110,9 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -21603,7 +21603,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -21623,33 +21623,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 327, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 326, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -23617,7 +23617,7 @@ ChildProcessError: Traceback (most recent call last): AssertionError: AWQ kernels could not be loaded. Please install them from https://github.com/casper-hansen/AutoAWQ_kernels ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, -4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemv-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,gemv,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -23637,33 +23637,33 @@ ChildProcessError: Traceback (most recent call last): return self.pretrained_model.generate(**inputs, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/utils/_contextlib.py"", line 115, in decorate_context return func(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1758, in generate + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 1914, in generate result = self._sample( - File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2397, in _sample + File ""/usr/local/lib/python3.10/dist-packages/transformers/generation/utils.py"", line 2651, in _sample outputs = self( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1164, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 1174, in forward outputs = self.model( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 968, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 978, in forward layer_outputs = decoder_layer( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 713, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 718, in forward hidden_states, self_attn_weights, present_key_value = self.self_attn( File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1541, in _call_impl return forward_call(*args, **kwargs) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 327, in forward + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/llama/modeling_llama.py"", line 326, in forward query_states = self.q_proj(hidden_states) File ""/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py"", line 1532, in _wrapped_call_impl return self._call_impl(*args, **kwargs) @@ -23963,7 +23963,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -23981,15 +23981,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -24355,7 +24355,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -24373,15 +24373,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -26352,7 +26352,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -26370,15 +26370,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -27772,7 +27772,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -27790,15 +27790,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -28325,7 +28325,7 @@ ValueError: GPTNeoXForCausalLM does not support an attention implementation thro ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, 4bit-awq-gemm-sdpa,pytorch,2.3.0+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,meta-llama/Llama-2-7b-hf,meta-llama/Llama-2-7b-hf,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.215-203.850.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.0,,0.30.1,,,,1.19.2,,,,0.11.1,,,,,,MB,1984.512,5480.382464,0.0,4833.93536,4503.282688,s,10,5.704772033691406,0.5704772033691407,0.0012601125399919193,0.5702384033203125,0.5718997436523438,0.5722093933105469,0.5724571130371093,"[0.5703561401367188, 0.57251904296875, 0.5679871215820312, 0.569779052734375, 0.569513427734375, 0.5698793334960938, 0.5701206665039062, 0.571189208984375, 0.5718309326171875, 0.5715971069335938]",tokens/s,448.7471164283303,kWh,6.715062223834757e-06,3.6795454622733808e-06,3.136956213266472e-05,4.1764169818772856e-05,tokens/kWh,6129656.140918401,MB,1986.158592,5480.382464,0.0,4833.93536,4688.699392,s,10,334.56283984375,33.456283984375,0.0052644785154184115,33.45499609375,33.46062109375,33.46519140625,33.46884765625,"[33.4560859375, 33.4513203125, 33.45960546875, 33.45390625, 33.45316015625, 33.46976171875, 33.4577109375, 33.45030078125, 33.45712109375, 33.4538671875]",tokens/s,1.883054317371969,kWh,0.0003949493847824173,0.00021646648750867824,0.0018252696453997421,0.0024366855176908373,tokens/kWh,25854.793137073728,,s,629,339.16590026855516,0.5392144678355401,0.06782032837407061,0.5309541625976563,0.5315696655273437,0.5319477294921875,1.101242978515625,"[0.5305211181640626, 0.5310648193359375, 0.5307863159179688, 0.53155224609375, 0.5311129760742187, 0.5313126220703125, 0.5308641357421875, 0.5310208129882813, 0.530787353515625, 0.5312348022460938, 0.5306480712890626, 0.5307863159179688, 0.5306419067382813, 0.5310637817382813, 0.530830322265625, 0.5309419555664062, 0.5309429931640625, 0.531325927734375, 0.5308753662109374, 0.5309122314453125, 0.5310187377929687, 0.5310218505859375, 0.5317243041992188, 0.5311610717773437, 0.5310167236328125, 0.5312348022460938, 0.530935791015625, 0.5312266235351563, 0.53100439453125, 0.5311580200195313, 0.5308682250976563, 0.5309685668945312, 0.5309716186523438, 0.5314590454101562, 0.5310986328125, 0.53151025390625, 0.5308231811523437, 0.5309214477539063, 0.5307422485351563, 0.5312634887695312, 0.5309501342773437, 0.5314273071289063, 0.530819091796875, 0.53110986328125, 0.5307781372070313, 0.53089892578125, 0.5306757202148438, 0.53097265625, 0.5308671875, 0.530881591796875, 0.5306766967773437, 0.53083544921875, 0.5309061279296875, 0.5309368286132813, 0.531083251953125, 0.5310986328125, 0.531441650390625, 0.531240966796875, 0.5315379028320313, 0.5314641723632813, 0.5310187377929687, 0.5314037475585938, 1.103942626953125, 0.5305426025390625, 0.5309706420898438, 0.530461669921875, 0.5309747314453125, 0.530756591796875, 0.5311068115234375, 0.5308692626953125, 0.5308150024414062, 0.5308845825195313, 0.5310484619140625, 0.5306859741210938, 0.5307606811523438, 0.5305077514648437, 0.5314058227539062, 0.53094091796875, 0.5307955322265625, 0.5311867065429687, 0.53100341796875, 0.5308580322265625, 0.5310903930664063, 0.5308118896484375, 0.5307975463867187, 0.5306583251953125, 0.5308969116210938, 0.530861083984375, 0.5308999633789062, 0.5306531982421875, 0.5309398803710937, 0.5309183959960937, 0.5308845825195313, 0.530808837890625, 0.5309910888671875, 0.5313074951171874, 0.5310607299804687, 0.530740234375, 0.5311702880859375, 0.53077197265625, 0.5312318115234375, 0.5309869384765625, 0.5307576293945313, 0.5307412719726563, 0.5308344116210938, 0.5307822265625, 0.5309531860351563, 0.5305784301757812, 0.5323253784179688, 0.5312071533203125, 0.5312880859375, 0.5308795166015625, 0.5310740356445313, 0.5312593994140625, 0.5309235229492187, 0.531114990234375, 0.5311170654296875, 0.5312634887695312, 0.5309481201171875, 0.5309020385742188, 0.5311262817382812, 0.5310023803710937, 0.5309122924804688, 0.5308917236328125, 0.5311027221679687, 1.100294189453125, 0.5306101684570312, 0.530904052734375, 0.5310392456054688, 0.5307822265625, 0.530640869140625, 0.5311344604492187, 0.5307207641601562, 0.531220458984375, 0.5305763549804687, 0.530935791015625, 0.5307678833007813, 0.5314918212890625, 0.5320335083007812, 0.5323519897460938, 0.5317181396484375, 0.5310802001953125, 0.5309747314453125, 0.5312337646484375, 0.5312235717773437, 0.5314109497070313, 0.5310474243164063, 0.5315625, 0.5313208618164063, 0.5311416015625, 0.531430419921875, 0.5316690063476562, 0.5314365234375, 0.531483642578125, 0.530713623046875, 0.5309296875, 0.5309389038085938, 0.5307412719726563, 0.5305579223632813, 0.530681884765625, 0.5306705932617187, 0.5307053833007812, 0.5305855712890625, 0.5307422485351563, 0.53066650390625, 0.5308958740234375, 0.532068359375, 0.532295654296875, 0.5322301635742187, 0.5322598266601563, 0.5318870849609375, 0.5309050903320313, 0.5312890625, 0.5309685668945312, 0.5306531982421875, 0.5310576782226563, 0.530798583984375, 0.5307260131835938, 0.5306234130859375, 0.5307207641601562, 0.5306265869140625, 0.5307678833007813, 0.530714599609375, 0.5308375244140625, 0.53097265625, 0.5309849853515625, 0.5310422973632812, 0.5310628051757813, 1.1018045654296875, 0.5307658081054687, 0.5312337646484375, 0.5306675415039063, 0.5308784790039063, 0.5308600463867188, 0.5310167236328125, 0.531009521484375, 0.5312532348632812, 0.5308733520507812, 0.5311190795898437, 0.5311856689453125, 0.5314058227539062, 0.5310842895507812, 0.5315277099609375, 0.5310382080078125, 0.5311743774414063, 0.5315983276367188, 0.5321942749023437, 0.5316853637695312, 0.5308467407226563, 0.5309767456054687, 0.530850830078125, 0.5307125854492187, 0.5308641357421875, 0.5307555541992187, 0.5308436279296875, 0.5307218017578125, 0.5308006591796876, 0.5305630493164063, 0.5307371215820312, 0.530661376953125, 0.5308436889648438, 0.5309583129882812, 0.5307883911132812, 0.5307330322265625, 0.5312798461914062, 0.5310320434570313, 0.5316075439453125, 0.5309869995117188, 0.5313341674804688, 0.5305620727539062, 0.5311702880859375, 0.5308067626953125, 0.5318553466796875, 0.5307586669921875, 0.5308836059570312, 0.5307012939453125, 0.5307760620117188, 0.5307177124023438, 0.5308969116210938, 0.530735107421875, 0.5311006469726562, 0.5306951904296875, 0.5307473754882812, 0.5309122314453125, 0.5312890625, 0.5314263305664062, 0.530976806640625, 0.530935791015625, 0.530862060546875, 0.530808837890625, 0.5308170166015626, 1.100686279296875, 0.5305743408203125, 0.5307484130859375, 0.53056103515625, 0.5306112060546875, 0.5306911010742188, 0.5307637939453125, 0.5306419067382813, 0.5308395385742187, 0.530756591796875, 0.5308426513671874, 0.5308016357421875, 0.5311057739257813, 0.5309327392578125, 0.5313269653320313, 0.5308795166015625, 0.531177490234375, 0.5310422973632812, 0.53136279296875, 0.5311201171875, 0.5312051391601562, 0.53100439453125, 0.5311467895507812, 0.5310556030273438, 0.531061767578125, 0.5310576782226563, 0.5312481079101562, 0.5311692504882812, 0.5314559936523438, 0.5308375244140625, 0.5311375122070312, 0.5306767578125, 0.530946044921875, 0.53065625, 0.5307238159179688, 0.5309081420898437, 0.5308159790039062, 0.5306972045898437, 0.53074951171875, 0.5306746215820313, 0.5307698974609375, 0.5306572875976563, 0.530976806640625, 0.5309541625976563, 0.5310791625976562, 0.531009521484375, 0.5310802001953125, 0.5311498413085938, 0.5309020385742188, 0.5308917846679687, 0.5309102172851563, 0.5309030151367188, 0.5314529418945313, 0.5309439697265625, 0.5307105102539063, 0.5309010009765625, 0.5310361328125, 0.5321871337890625, 0.5309573364257812, 0.5309470825195313, 0.5312696533203125, 0.5320478515625, 0.5322874755859375, 1.10145947265625, 0.5313740844726562, 0.5319639282226563, 0.53125732421875, 0.5318215942382812, 0.5319761962890625, 0.5323817138671875, 0.5318543090820312, 0.5324503173828125, 0.53193115234375, 0.53076171875, 0.5305753784179688, 0.5308323974609375, 0.530567138671875, 0.5308436279296875, 0.5306286010742187, 0.5307053833007812, 0.5305753784179688, 0.53100341796875, 0.5306358032226562, 0.5310361328125, 0.5309552612304688, 0.5309389038085938, 0.53064501953125, 0.53108837890625, 0.5308590087890624, 0.5309389038085938, 0.5310259399414062, 0.5312952270507812, 0.5310709838867187, 0.5311764526367188, 0.5312491455078125, 0.5314150390625, 0.5312532348632812, 0.531252197265625, 0.5313546142578125, 0.531367919921875, 0.5319331665039062, 0.5335838623046875, 0.5311641845703124, 0.531441650390625, 0.5313925170898437, 0.5322434692382813, 0.5322864379882812, 0.5321441040039062, 0.532115478515625, 0.531388427734375, 0.5306961669921875, 0.53096240234375, 0.5312880859375, 0.53085693359375, 0.5306890869140625, 0.5307586059570313, 0.53071875, 0.5307207641601562, 0.53096240234375, 0.5315604248046875, 0.5309337768554687, 0.5310587158203125, 0.5309173583984375, 0.53144677734375, 0.530819091796875, 0.5310504760742187, 1.1015126953125, 0.5308446655273438, 0.53172021484375, 0.5307381591796875, 0.5307678833007813, 0.5306972045898437, 0.5308292846679687, 0.5307166748046875, 0.530808837890625, 0.5306388549804687, 0.5306736450195313, 0.5309235229492187, 0.5309235229492187, 0.53070849609375, 0.5309276123046875, 0.5306429443359375, 0.5307852783203125, 0.5307914428710937, 0.5309736938476562, 0.5309429931640625, 0.5308538818359375, 0.530862060546875, 0.530893798828125, 0.5308037109375, 0.530777099609375, 0.5308395385742187, 0.5308600463867188, 0.5307258911132813, 0.5308436279296875, 0.530629638671875, 0.531146728515625, 0.5308661499023437, 0.531294189453125, 0.5314826049804687, 0.5312481079101562, 0.53166796875, 0.5312839965820313, 0.53075048828125, 0.530819091796875, 0.5311170654296875, 0.5316137084960938, 0.5317027587890625, 0.5319608154296875, 0.5318799438476562, 0.5317857055664063, 0.531937255859375, 0.5316761474609375, 0.5316976928710937, 0.5317744750976563, 0.5308323974609375, 0.531051513671875, 0.5311907958984375, 0.5314478149414062, 0.531346435546875, 0.5310218505859375, 0.5310238647460938, 0.5312337646484375, 0.5313167114257813, 0.5308651733398437, 0.530703369140625, 0.5307924194335938, 0.5308026733398438, 0.5309132690429688, 1.1025264892578126, 0.53065625, 0.5309061279296875, 0.5318810424804687, 0.5307862548828125, 0.530639892578125, 0.53081396484375, 0.5309368286132813, 0.53110888671875, 0.5308538818359375, 0.5307095336914063, 0.5305200805664062, 0.5307893676757812, 0.5306808471679687, 0.530820068359375, 0.5304883422851563, 0.5308170166015626, 0.5304985961914063, 0.5307473754882812, 0.5305302734375, 0.5306531982421875, 0.530513916015625, 0.5310648193359375, 0.5307095336914063, 0.5307955322265625, 0.5309153442382812, 0.5314549560546875, 0.5309439697265625, 0.5312911376953126, 0.5316034545898437, 0.5309531860351563, 0.5309869995117188, 0.5310361328125, 0.5308016357421875, 0.5307443237304688, 0.5306705932617187, 0.5307801513671875, 0.530608154296875, 0.530639892578125, 0.5306634521484375, 0.5308221435546875, 0.5308283081054688, 0.5309696044921876, 0.5313065185546875, 0.5313300170898437, 0.5313269653320313, 0.5309522094726562, 0.531056640625, 0.5314488525390625, 0.5310975952148438, 0.5314590454101562, 0.53098291015625, 0.5310863647460937, 0.5309696044921876, 0.53103515625, 0.5311734008789063, 0.530808837890625, 0.530746337890625, 0.53139453125, 0.5307208251953125, 0.5316126708984374, 0.5313197631835938, 0.5314866943359375, 1.1037603759765624, 0.531114013671875, 0.5313157348632812, 0.5316669311523438, 0.5312061157226563, 0.5314129638671875, 0.5316454467773437, 0.531177490234375, 0.5316536865234375, 0.530905029296875, 0.5309481201171875, 0.5308477172851562, 0.5310320434570313, 0.5309378662109375, 0.5309757690429687, 0.5313228759765625, 0.5311948852539062, 0.5310637817382813, 0.5309552612304688, 0.5314559936523438, 0.5309481201171875, 0.5310003051757812, 0.5309910888671875, 0.5306224365234375, 0.5306593017578125, 0.5306255493164063, 0.5308753662109374, 0.5307627563476562, 0.5307576293945313, 0.530713623046875, 0.53089892578125, 0.530850830078125, 0.5307238159179688, 0.5306982421875, 0.5311590576171875, 0.530724853515625, 0.5312921752929688, 0.5312542724609375, 0.5312317504882812, 0.5316403198242188, 0.5309900512695312, 0.5310330810546875, 0.5311273193359375, 0.5309204711914063, 0.53094091796875, 0.5306009521484375, 0.530967529296875, 0.5310392456054688, 0.5308815307617187, 0.5307689208984375, 0.5308272705078125, 0.5306911010742188, 0.5308804931640625, 0.5311528930664062, 0.5310679321289062, 0.5312716674804687, 0.531794921875, 0.531114013671875, 0.5313935546875, 0.5312348022460938, 0.5312020263671875, 0.5310812377929688, 0.5309931640625, 1.102993408203125, 0.5305927124023437, 0.53087744140625, 0.530703369140625, 0.5310504760742187, 0.5310187377929687, 0.5312553100585937, 0.5311918334960938, 0.5312839965820313, 0.5307781372070313, 0.5309859619140626, 0.5311580200195313, 0.5308999633789062, 0.5308969116210938, 0.5311795043945312, 0.5306071166992188, 0.5307647705078125, 0.5309890747070313, 0.5307586669921875, 0.5306500854492188, 0.530893798828125, 0.530766845703125, 0.5308999633789062, 0.5307760620117188, 0.5306542358398437, 0.530555908203125, 0.5319547119140625, 0.53079345703125, 0.5315061645507813, 0.5311846313476563, 0.5316218872070313, 0.5312348022460938, 0.5321912231445313, 0.5312000122070313, 0.531431396484375, 0.5311488037109375, 0.5308743896484375, 0.530819091796875, 0.5310812377929688, 0.5307105102539063, 0.53081396484375, 0.5309736938476562, 0.53100439453125, 0.5308211059570312, 0.5308876953125, 0.5309081420898437, 0.5309849853515625, 0.5307340698242188, 0.53097265625, 0.5311795043945312, 0.5309010009765625, 0.5311529541015625, 0.5311456909179687, 0.5306522216796875, 0.5312962036132812, 0.5308641357421875, 0.5309163818359375, 0.5308590087890624, 0.5311928100585938, 0.5311190795898437, 0.5309706420898438, 0.530935791015625, 0.531294189453125]",tokens/s,1.8545496451794004,,,,, -4bit-awq-gemm-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-gemm-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,gemm,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -28343,13 +28343,13 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3620, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3704, in from_pretrained config = cls._autoset_attn_implementation( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1478, in _autoset_attn_implementation + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1490, in _autoset_attn_implementation config = cls._check_and_enable_sdpa( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1644, in _check_and_enable_sdpa + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1656, in _check_and_enable_sdpa raise ValueError( ValueError: DeciLMForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` @@ -31571,7 +31571,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -31589,13 +31589,13 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3620, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3704, in from_pretrained config = cls._autoset_attn_implementation( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1478, in _autoset_attn_implementation + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1490, in _autoset_attn_implementation config = cls._check_and_enable_sdpa( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1644, in _check_and_enable_sdpa + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 1656, in _check_and_enable_sdpa raise ValueError( ValueError: DeciLMForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` @@ -33496,7 +33496,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: OPTForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -33514,15 +33514,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -34861,7 +34861,7 @@ ChildProcessError: Traceback (most recent call last): ValueError: GPTNeoXForCausalLM does not support an attention implementation through torch.nn.functional.scaled_dot_product_attention yet. Please request the support for this architecture: https://github.com/huggingface/transformers/issues/28005. If you believe this error is a bug, please open an issue in Transformers GitHub repository and load your model with the argument `attn_implementation=""eager""` meanwhile. Example: `model = AutoModel.from_pretrained(""openai/whisper-tiny"", attn_implementation=""eager"")` ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-sdpa,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,sdpa,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -34879,15 +34879,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -35515,7 +35515,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -35533,15 +35533,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 558, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 559, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -37494,7 +37494,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -37512,15 +37512,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -38896,7 +38896,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exl_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1,64,1 -4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v1-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -38914,15 +38914,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 461, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 462, in post_init_awq_exllama_modules model = exllama_post_init(model) File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllama.py"", line 144, in exllama_post_init submodule.post_init() @@ -39552,7 +39552,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -39560,7 +39560,7 @@ NameError: name 'exlv2_ext' is not defined File ""/workspace/optimum_benchmark/launchers/process/launcher.py"", line 66, in launch raise ChildProcessError(response[""traceback""]) ChildProcessError: Traceback (most recent call last): - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 612, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 613, in resolve_trust_remote_code answer = input( EOFError: EOF when reading a line @@ -39577,9 +39577,9 @@ Traceback (most recent call last): self.create_no_weights_model() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 210, in create_no_weights_model meta_model = self.automodel_class.from_config(self.pretrained_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 418, in from_config + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 419, in from_config trust_remote_code = resolve_trust_remote_code( - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 625, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 626, in resolve_trust_remote_code raise ValueError( ValueError: The repository for Deci/DeciLM-7B contains custom code which must be executed to correctly load the model. You can inspect the repository content at https://hf.co/Deci/DeciLM-7B. Please pass the argument `trust_remote_code=True` to allow custom code to be run. @@ -41541,7 +41541,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -41559,15 +41559,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device]) @@ -42949,7 +42949,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-flash_attention_2,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,flash_attention_2,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -42967,15 +42967,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device]) @@ -43219,7 +43219,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciCoder-1b,Deci/DeciCoder-1b,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -43227,7 +43227,7 @@ NameError: name 'exlv2_ext' is not defined File ""/workspace/optimum_benchmark/launchers/process/launcher.py"", line 66, in launch raise ChildProcessError(response[""traceback""]) ChildProcessError: Traceback (most recent call last): - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 612, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 613, in resolve_trust_remote_code answer = input( EOFError: EOF when reading a line @@ -43244,9 +43244,9 @@ Traceback (most recent call last): self.create_no_weights_model() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 210, in create_no_weights_model meta_model = self.automodel_class.from_config(self.pretrained_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 418, in from_config + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 419, in from_config trust_remote_code = resolve_trust_remote_code( - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 625, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 626, in resolve_trust_remote_code raise ValueError( ValueError: The repository for Deci/DeciCoder-1b contains custom code which must be executed to correctly load the model. You can inspect the repository content at https://hf.co/Deci/DeciCoder-1b. Please pass the argument `trust_remote_code=True` to allow custom code to be run. @@ -43609,7 +43609,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,Deci/DeciLM-7B,Deci/DeciLM-7B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -43617,7 +43617,7 @@ NameError: name 'exlv2_ext' is not defined File ""/workspace/optimum_benchmark/launchers/process/launcher.py"", line 66, in launch raise ChildProcessError(response[""traceback""]) ChildProcessError: Traceback (most recent call last): - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 612, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 613, in resolve_trust_remote_code answer = input( EOFError: EOF when reading a line @@ -43634,9 +43634,9 @@ Traceback (most recent call last): self.create_no_weights_model() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 210, in create_no_weights_model meta_model = self.automodel_class.from_config(self.pretrained_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 418, in from_config + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 419, in from_config trust_remote_code = resolve_trust_remote_code( - File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 625, in resolve_trust_remote_code + File ""/usr/local/lib/python3.10/dist-packages/transformers/dynamic_module_utils.py"", line 626, in resolve_trust_remote_code raise ValueError( ValueError: The repository for Deci/DeciLM-7B contains custom code which must be executed to correctly load the model. You can inspect the repository content at https://hf.co/Deci/DeciLM-7B. Please pass the argument `trust_remote_code=True` to allow custom code to be run. @@ -45610,7 +45610,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-34B,01-ai/Yi-34B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -45628,15 +45628,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device]) @@ -47030,7 +47030,7 @@ ChildProcessError: Traceback (most recent call last): NameError: name 'exlv2_ext' is not defined ",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,2,64,1 -4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.41.2,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): +4bit-awq-exllama-v2-eager,pytorch,2.3.1+cu121,optimum_benchmark.backends.pytorch.backend.PyTorchBackend,text-generation,transformers,01-ai/Yi-6B,01-ai/Yi-6B,cuda,0,42,,,True,True,True,True,,float16,True,False,,eager,,False,,False,forward,awq,4,exllama,False,,inference,optimum_benchmark.scenarios.inference.scenario.InferenceScenario,10,10,10,1,2,256,,True,True,True,64,64,process,optimum_benchmark.launchers.process.launcher.ProcessLauncher,True,kill,False,spawn, AMD EPYC 7R32,16,66697.29792,Linux,x86_64,Linux-5.10.218-208.862.amzn2.x86_64-x86_64-with-glibc2.35,x86_64,3.10.12,['NVIDIA A10G'],1,24146608128,0.2.1,,4.42.1,,0.31.0,,,,1.20.0,,,,0.11.1,,"Traceback (most recent call last): File ""/workspace/llm_perf/update_llm_perf_cuda_pytorch.py"", line 148, in benchmark_cuda_pytorch benchmark_report = Benchmark.launch(benchmark_config) File ""/workspace/optimum_benchmark/benchmark/base.py"", line 47, in launch @@ -47048,15 +47048,15 @@ ChildProcessError: Traceback (most recent call last): self.load_model_from_pretrained() File ""/workspace/optimum_benchmark/backends/pytorch/backend.py"", line 169, in load_model_from_pretrained self.pretrained_model = self.automodel_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 563, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/models/auto/auto_factory.py"", line 564, in from_pretrained return model_class.from_pretrained( - File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3823, in from_pretrained + File ""/usr/local/lib/python3.10/dist-packages/transformers/modeling_utils.py"", line 3907, in from_pretrained hf_quantizer.postprocess_model(model) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/base.py"", line 195, in postprocess_model return self._process_model_after_weight_loading(model, **kwargs) File ""/usr/local/lib/python3.10/dist-packages/transformers/quantizers/quantizer_awq.py"", line 107, in _process_model_after_weight_loading model = post_init_awq_exllama_modules(model, self.quantization_config.exllama_config) - File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 465, in post_init_awq_exllama_modules + File ""/usr/local/lib/python3.10/dist-packages/transformers/integrations/awq.py"", line 466, in post_init_awq_exllama_modules model = exllamav2_post_init( File ""/usr/local/lib/python3.10/dist-packages/awq/modules/linear/exllamav2.py"", line 198, in exllamav2_post_init submodule.post_init(scratch_space=model.scratch_spaces[device])