I have searched the existing issues and discussions.
Problem Description
Every time a file is uploaded in a new chat, it always goes through content extraction. This process can be slow and the result can be incorrect (a pdf often can be seen as an empty). When a file is uploaded and seen as a document, the prompt is also forced to change to RAG template, which often is not the desired case.
Desired Solution you'd like
Separate a file upload and document upload.
If a file needs to be RAG, it can go through the content extraction and RAG template as before. If it is uploaded as a file, and the model supports file input, skip RAG and simply send the file to the backend model.
Alternatives Considered
No response
Additional Context
No response
Originally created by @CuriousLocky on GitHub (Jun 26, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/15338
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
### Problem Description
Every time a file is uploaded in a new chat, it always goes through content extraction. This process can be slow and the result can be incorrect (a pdf often can be seen as an empty). When a file is uploaded and seen as a document, the prompt is also forced to change to RAG template, which often is not the desired case.
### Desired Solution you'd like
Separate a file upload and document upload.
If a file needs to be RAG, it can go through the content extraction and RAG template as before. If it is uploaded as a file, and the model supports file input, skip RAG and simply send the file to the backend model.
### Alternatives Considered
_No response_
### Additional Context
_No response_
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 @CuriousLocky on GitHub (Jun 26, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/15338
Check Existing Issues
Problem Description
Every time a file is uploaded in a new chat, it always goes through content extraction. This process can be slow and the result can be incorrect (a pdf often can be seen as an empty). When a file is uploaded and seen as a document, the prompt is also forced to change to RAG template, which often is not the desired case.
Desired Solution you'd like
Separate a file upload and document upload.
If a file needs to be RAG, it can go through the content extraction and RAG template as before. If it is uploaded as a file, and the model supports file input, skip RAG and simply send the file to the backend model.
Alternatives Considered
No response
Additional Context
No response