mirror of
https://github.com/open-webui/open-webui.git
synced 2026-05-07 11:28:35 -05:00
[GH-ISSUE #23686] issue: Missing document context in channel messages #35575
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @carpemonf on GitHub (Apr 13, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/23686
Check Existing Issues
Installation Method
Pip Install
Open WebUI Version
0.8.12
Ollama Version (if applicable)
No response
Operating System
Ubuntu 24.04
Browser (if applicable)
No response
Confirmation
README.md.Expected Behavior
In channels, questions directed to the AI model should have access to the full conversation context. I guess this includes previous messages, as well as any uploaded documents or content shared in the channel.
After uploading a document or pasting its content, asking a question (and mentioning the model) should produce a response that clearly uses that context, rather than a generic answer.
Actual Behavior
In channel conversations, the model does not appear to use the provided document context. After uploading a document (PDF) and asking a question about it (while mentioning the model), the response is generic and not grounded in the document.
This also happens when the document content is manually pasted into the channel, the model still responds as if no additional context was supplied.
This does not occur in regular Chats, where uploading the same PDF and running the same test works as expected.
Steps to Reproduce
pip install open-webui==0.8.12open-webui serveLogs & Screenshots
Chat usage
Logs shows that Open WebUI uses
query_docand sends related context toollama:Channel usage
There are no
query_doccalling from channels, and Ollama receives the following prompt instead:Additional Information
No response
@tjbck commented on GitHub (Apr 14, 2026):
#8050