I have searched the existing issues and discussions.
Problem Description
Currently, the citation/quote feature in Open WebUI only displays limited (none) information about which chunks from the knowledge base were used to generate an answer. This makes it difficult for users to verify the detailed source context, especially in cases where multiple knowledge base entries or larger documents are referenced. Transparency is crucial for trust and validation in Retrieval Augmented Generation (RAG) workflows.
Desired Solution you'd like
Please enhance the citation/quote functionality so that all relevant chunks used for generating any particular response are displayed in full. Ideally, after receiving an answer, users should see a clearly organized list or expandable view of all source chunks (text passages, paragraphs, document references, etc.) that contributed to the final model output. Navigating directly to those chunks or previewing them in context (optionally with highlighting) would make the feature significantly more valuable.
Some greate examples:
Perplexity
NotebookLM
Alternatives Considered
No response
Additional Context
No response
Originally created by @SchneiderSam on GitHub (Aug 25, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/16894
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
### Problem Description
Currently, the citation/quote feature in Open WebUI only displays limited (none) information about which chunks from the knowledge base were used to generate an answer. This makes it difficult for users to verify the detailed source context, especially in cases where multiple knowledge base entries or larger documents are referenced. Transparency is crucial for trust and validation in Retrieval Augmented Generation (RAG) workflows.
### Desired Solution you'd like
Please enhance the citation/quote functionality so that all relevant chunks used for generating any particular response are displayed in full. Ideally, after receiving an answer, users should see a clearly organized list or expandable view of all source chunks (text passages, paragraphs, document references, etc.) that contributed to the final model output. Navigating directly to those chunks or previewing them in context (optionally with highlighting) would make the feature significantly more valuable.
**Some greate examples:**
Perplexity
<img width="1204" height="367" alt="Image" src="https://github.com/user-attachments/assets/b135d4a5-5b2d-4aa5-88a2-510fa41b2eb0" />
NotebookLM
<img width="1491" height="933" alt="Image" src="https://github.com/user-attachments/assets/7b017cff-871a-4596-92ef-0b090e0654be" />
### Alternatives Considered
_No response_
### Additional Context
_No response_
You can click on the inline citations for the retrieved text.
where? :-)
<!-- gh-comment-id:3234260373 -->
@SchneiderSam commented on GitHub (Aug 28, 2025):
> You can click on the inline citations for the retrieved text.
where? :-)
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @SchneiderSam on GitHub (Aug 25, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/16894
Check Existing Issues
Problem Description
Currently, the citation/quote feature in Open WebUI only displays limited (none) information about which chunks from the knowledge base were used to generate an answer. This makes it difficult for users to verify the detailed source context, especially in cases where multiple knowledge base entries or larger documents are referenced. Transparency is crucial for trust and validation in Retrieval Augmented Generation (RAG) workflows.
Desired Solution you'd like
Please enhance the citation/quote functionality so that all relevant chunks used for generating any particular response are displayed in full. Ideally, after receiving an answer, users should see a clearly organized list or expandable view of all source chunks (text passages, paragraphs, document references, etc.) that contributed to the final model output. Navigating directly to those chunks or previewing them in context (optionally with highlighting) would make the feature significantly more valuable.
Some greate examples:
Perplexity
NotebookLM
Alternatives Considered
No response
Additional Context
No response
@tjbck commented on GitHub (Aug 28, 2025):
You can click on the inline citations for the retrieved text.
@SchneiderSam commented on GitHub (Aug 28, 2025):
where? :-)
@SchneiderSam commented on GitHub (Aug 29, 2025):
ok got it