The agent cannot answer the question.

[ganzhipeng111]

回忆已完成
正在思考中...
💥 运行出错： Error in generating model output: Error code: 400 - {'error': {'message': 'Failed to deserialize the JSON body into the target type: messages[2]: invalid type: sequence, expected a string at line 1 column 12754', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
💥 运行出错： Error in generating model output: Error code: 400 - {'error': {'message': 'Failed to deserialize the JSON body into the target type: messages[2]: invalid type: sequence, expected a string at line 1 column 12754', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
💥 运行出错： Error in generating model output: Error code: 400 - {'error': {'message': 'Failed to deserialize the JSON body into the target type: messages[2]: invalid type: sequence, expected a string at line 1 column 12754', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
💥 运行出错： Error in generating model output: Error code: 400 - {'error': {'message': 'Failed to deserialize the JSON body into the target type: messages[2]: invalid type: sequence, expected a string at line 1 column 12754', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}
💥 运行出错： Error in generating model output: Error code: 400 - {'error': {'message': 'Failed to deserialize the JSON body into the target type: messages[2]: invalid type: sequence, expected a string at line 1 column 12754', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_request_error'}}

[ganzhipeng111]

It seems to be related to the tools integrated with the agent. When integrating local tools, the error shown in the screenshot occurs; whereas when integrating the Nexent tool, the agent can answer questions normally.

[WMC001]

Thank you for your observation. We would like to know how the local tool you mentioned is integrated.

[ganzhipeng111]

集成这个单词出现是翻译出问题，不用关心。我就是配置了非常简单的工具，就是工具就调用了本地知识库与nexent，就没了。我展示一下我的工具配置信息。就配置了两个，如下图所示

[Jasonxia007]

This error message seems have something to do with Deepseek model output. Maybe you can try to use another LLM or Deepseek service from a third-party platform to avoid this problem.

[ganzhipeng111]

此错误消息似乎与 Deepseek 模型输出有关。也许你可以尝试使用第三方平台的其他 LLM 或 Deepseek 服务来避免这个问题。

Following your advice, I have switched to a different large language model – the QwQ-32B model developed by Guiji Liudong. This model can now access the local knowledge base, but the local knowledge base still fails to import PDF documents into its library.

[Phinease]

Can you please attach screenshot of nexent or nexent-data-process container's log? Thanks

[ganzhipeng111]

ok，this is nexent-data-process container's log
nexent-data-process container's log.docx