You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I love the search function and it makes the chat feel so much more accurate! I use it mainly as a direct ChatGPT replacment, using code models when needed or normal models for chat.
Can we have the option to upload images/pdfs/other files to the chat? the images could be integrated by clip/blip, and the PDF or text files could just be added to the context or summarized and then added?
It would be awesome to have! Thank you for all the work made into this project
The text was updated successfully, but these errors were encountered:
I'm not sure if this is really the right place to comment, but I just wanted to document a similar feature request/use case.
We have an existing RAG engine which is highly customised to our particular use case. Currently this is behind a streamlit frontend, but we'd like to investigate moving this to chat-ui.
Currently we could implement the OpenAI protocol and provide this RAG engine as a model, but we'd have no way of configuring additional switches (for example which data sources to enable) or nicely rendering things like citations.
I'm not sure what this would look like for chat-ui, or if it's out of scope, but I wanted to put the use case out there.
I love the search function and it makes the chat feel so much more accurate! I use it mainly as a direct ChatGPT replacment, using code models when needed or normal models for chat.
Can we have the option to upload images/pdfs/other files to the chat? the images could be integrated by clip/blip, and the PDF or text files could just be added to the context or summarized and then added?
It would be awesome to have! Thank you for all the work made into this project
The text was updated successfully, but these errors were encountered: