search

infiniflow / ragflow

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
https://ragflow.io
Apache License 2.0
10.08k stars 974 forks source link

[Bug]: the current embedding module is not user configured at knowledgebase when when update/create new chunk. #1183

Closed chinamerp closed 1 week ago

chinamerp commented 1 week ago

Is there an existing issue for the same bug?

Branch name

main

Commit ID

1eb4caf02a9fdb1f3c5a7e8f304163d8823bfa5c

Other environment information

No response

Actual behavior

It use the global embedding module

Expected behavior

using user configured at knowledgebase

Steps to reproduce

create new chunk or update a chunk

Additional information

No response

KevinHuSh commented 1 week ago

Which version of image you used? I remembered that it's fixed in the dev version of docker image.

chinamerp commented 1 week ago

see the code: it use global setting, not user configured at kb embd_mdl = TenantLLMService.model_instance( tenant_id, LLMType.EMBEDDING.value)

KevinHuSh commented 1 week ago

tenant_id is used for fetching the user configured model settings.

chinamerp commented 1 week ago

那知识库里配置的嵌入模型的作用是什么呢？查询和创建不使用同一个嵌入模型？