[Bug]: ERROR: Error code: 400 - {'detail': '[address=0.0.0.0:51672, pid=57224] Model not found, uid: qwen1.5-chat-ieJFufN9-1-0'}

[ipv6next]

Is there an existing issue for the same bug?

• [X] I have checked the existing issues.

Branch name

main

Commit ID

1234

Other environment information

LLM : qwen-1.5 本地部署
embe: large-zh 本地

通过 xinference 加载

Actual behavior

正常用LLM，可以聊天。 但是调用知识库时候，报错 ERROR: Error code: 400 - {'detail': '[address=0.0.0.0:51672, pid=57224] Model not found, uid: qwen1.5-chat-ieJFufN9-1-0'}

Expected behavior

No response

Steps to reproduce

正常用LLM，可以聊天。
但是调用知识库时候，报错
ERROR: Error code: 400 - {'detail': '[address=0.0.0.0:51672, pid=57224] Model not found, uid: qwen1.5-chat-ieJFufN9-1-0'}

Additional information

No response

[KevinHuSh]

Is this helpful? https://github.com/infiniflow/ragflow/blob/main/docs/faq.md#414-hint--102--fail-to-access-model--connection-error

[lawsonX]

@KevinHuSh Do we have any conclusions about the issue? I have encountered similar problem in ragflow v0.8 ，with xinferece local qwen model.

Error code: 400 - {'detail': '[address=0.0.0.0:34838, pid=2945997] Model not found, uid: qwen1.5-chat-1-0'}

based on the error, the model UID is wrong with the suffix "-1-0", which is not from my configuration, it should not exists.

This BUG happens occasionally only with API usage, and will return to normal by itself. So weird