【问题描述】
单卡推理,mindie服务化拉起失败,报错:"error":"Failed to get engine response.","error_type":"Incomplete Generation"
debug日志:
[2025-03-31 10:00:35.410282] [316] [281450121064800] [llm] [ERROR] [model.cpp:348] [MIE05E000002] Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
[2025-03-31 10:00:35,410] [316] [281450121064800] [llm] [ERROR] [logging.py-51] : Unknown exception: Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
[2025-03-31 10:00:35.031+0800] [316] [281459163787616] [mindie-server] [ERROR] [standard_model.py:142] : [Model] >>>Execute type:2, Exception:Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 133, in execute
return self._decode(requests)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 231, in _decode
return self.__generate(requests, is_prefill=False, is_mix=False)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 357, in __generate
request_tokens_np, eos_np, truncation_indices = self.__handle_requests(requests, is_prefill, is_mix)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 580, in __handle_requests
request_tokens_np, eos_np, _, truncation_indices = self.generator.generate_token(metadata)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 190, in generate_token
raise e
【镜像版本】
MindIE:2.0.T3.1-800I-A2-py311-openeuler24.03-lts
【芯片类型】
910
【模型】Qwen2系列,Qwen2.5-0.5B-Instruct
【问题描述】
单卡推理,mindie服务化拉起失败,报错:"error":"Failed to get engine response.","error_type":"Incomplete Generation"
debug日志:
[2025-03-31 10:00:35.410282] [316] [281450121064800] [llm] [ERROR] [model.cpp:348] [MIE05E000002] Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
[2025-03-31 10:00:35,410] [316] [281450121064800] [llm] [ERROR] [logging.py-51] : Unknown exception: Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
[2025-03-31 10:00:35.031+0800] [316] [281459163787616] [mindie-server] [ERROR] [standard_model.py:142] : [Model] >>>Execute type:2, Exception:Execute fail, enable log: export ASDOPS_LOG_LEVEL=ERROR, export ASDOPS_LOG_TO_STDOUT=1 to findthe first error. For more details, see the MindIE official document.
Traceback (most recent call last):
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 133, in execute
return self._decode(requests)
^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 231, in _decode
return self.__generate(requests, is_prefill=False, is_mix=False)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 357, in __generate
request_tokens_np, eos_np, truncation_indices = self.__handle_requests(requests, is_prefill, is_mix)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/model_wrapper/standard_model.py", line 580, in __handle_requests
request_tokens_np, eos_np, _, truncation_indices = self.generator.generate_token(metadata)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.11/site-packages/mindie_llm/text_generator/generator.py", line 190, in generate_token
raise e
【镜像版本】
MindIE:2.0.T3.1-800I-A2-py311-openeuler24.03-lts
【芯片类型】
910
【模型】Qwen2系列,Qwen2.5-0.5B-Instruct