问题描述：

用MindIE拉glm4-9b-chat模型，报错如下：

LLMInferEngine failed to init LLMInferModels
ERR: Failed to init endpoint! Please check the service log or console output.
Killed

问题定位

日志信息太少，多打些，重新测试

#MindIE Service+LLM日志 
export MINDIE_LOG_TO_STDOUT=true  #输入到屏幕
export MINDIE_LOG_LEVEL="debug"

#CANN日志收集
export ASCEND_GLOBAL_LOG_LEVEL=0  #debug级别
export ASCEND_SLOG_PRINT_TO_STDOUT=1  #输入到屏幕

#加速库日志 |默认输出路径为 
export ATB_LOG_LEVEL=DEBUG
export ATB_LOG_TO_STDOUT=1 #输入到屏幕

#算子库日志 |默认输出路径 ~/atb/log
export ASDOPS_LOG_LEVEL=DEBUG
export ASDOPS_LOG_TO_STDOUT=1 #输入到屏幕

报错如下：

Failed to get vocab size from tokenizer wrapper with exception: ValueError: safe_get_tokenizer_from_pretrained failed. Please check the input parameters model_path and kwargs. If the input parameters are valid and the required files exist in model_path, make sure the folder's owner has execute permission. Otherwise, please check the function stack for detailed exception information.

详细信息如下：

[WARNING] APP(1146,mindieservice_daemon):2025-03-19-19:33:36.595.960 [log_inner.cpp:79]1147 build/CMakeFiles/torch_npu.dir/compiler_depend.ts:get_default_custom_lib_path:118: "[PTA]:"config.ini is not exists""
[INFO] APP(1146,mindieservice_daemon):2025-03-19-19:33:36.596.099 [log_inner.cpp:82]1147 build/CMakeFiles/torch_npu.dir/compiler_depend.ts:GetDispatchTimeout:522: "[PTA]:"set dispatchTimeout_ 600 s.""
[2025-03-19 19:33:37.810928] [1146] [281472789114816] [llm] [ERROR] [model_deploy_config.cpp:152] Failed to get vocab size from tokenizer wrapper with exception: ValueError: safe_get_tokenizer_from_pretrained failed. Please check the input parameters model_path and kwargs. If the input parameters are valid and the required files exist in model_path, make sure the folder's owner has execute permission. Otherwise, please check the function stack for detailed exception information.

At:
  /usr/local/Ascend/atb-models/atb_llm/models/base/model_utils.py(207): wrapper
  /usr/local/Ascend/atb-models/atb_llm/models/base/router.py(323): get_tokenizer
  /usr/local/Ascend/atb-models/atb_llm/models/base/router.py(151): tokenizer
  /usr/local/Ascend/atb-models/atb_llm/runner/tokenizer_wrapper.py(20): __init__
  /usr/local/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/atb/atb_tokenizer_wrapper.py(8): __init__
  /usr/local/lib/python3.11/site-packages/mindie_llm/modeling/model_wrapper/__init__.py(28): get_tokenizer_wrapper

LLMInferEngine failed to init LLMInferModels
[2025-03-19 19:33:37.811+08:00] [1146] [ascend] [system] [server] [start engine(mindieservice_llm_engine)] [fail]
[2025-03-19 19:33:37.811+08:00] [1146] [1147] [mindie-server] [ERROR] [infer_backend_manager.cpp:236] : [MIE04E06011C] [infer_backend_manager] Failed to init engine: mindieservice_llm_engine
[2025-03-19 19:33:37.811+08:00] [1146] [1147] [mindie-server] [ERROR] [endpoint.cpp:53] : [MIE04E02011C] [endpoint] Failed to init engine! Please check in the mindservice.log, pythonlog.log or console output.
[2025-03-19 19:33:37.811+08:00] [1146] [ascend] [system] [server] [start mindie server] [fail]
[INFO] APP(1146,mindieservice_daemon):2025-03-19-19:33:47.812.236 [log_inner.cpp:82]1147 torch_npu/csrc/InitNpuBindings.cpp:THPModule_npu_shutdown_synchronize:87: "[PTA]:"NPU shutdown synchronize begin.""
[INFO] APP(1146,mindieservice_daemon):2025-03-19-19:33:47.812.300 [log_inner.cpp:82]1147 torch_npu/csrc/InitNpuBindings.cpp:THPModule_npu_shutdown:57: "[PTA]:"NPU shutdown begin.""
[2025-03-19 19:33:48.608+08:00] [1146] [ascend] [system] [server] [start endpoint] [fail]
[2025-03-19 19:33:48.609+08:00] [1146] [ascend] [system] [server] [start mindie server] [fail]
ERR: Failed to init endpoint! Please check the service log or console output.
[2025-03-19 19:33:48.610+08:00] [1146] [1147] [mindie-server] [INFO] [llm_daemon.cpp:142] : [daemon] ERR: Failed to init endpoint! Please check the service log or console output.
Killed

解决办法：

修改mindie配置文件，trustRemoteCode字段从false改为true：

vim /usr/local/Ascend/mindie/latest/mindie-service/conf/config.json