ParisNeo / lollms-webui

Lord of Large Language Models Web User Interface
https://parisneo.github.io/lollms-webui/
Apache License 2.0
4.11k stars 522 forks source link

AI not able to read documents uploaded #496

Open ba2512005 opened 5 months ago

ba2512005 commented 5 months ago

Expected Behavior

When I upload a document, I should be able to refer to it in the chat and prompt the AI to perform tasks with the contents.

Current Behavior

I upload a document and the AI has no idea that I uploaded it and cannot see it.

Steps to Reproduce

Select Exllama binding, select model. Upload document, ask AI about the document uploaded.

Possible Solution

Perhaps the document is not being parsed properly or loaded into context properly

Context

Windows 11, exllama binding

ParisNeo commented 5 months ago

Hi, what kind of document did you send to the AI?

ba2512005 commented 5 months ago

Docx

On Sat, Feb 24, 2024 at 3:43 AM Saifeddine ALOUI @.***> wrote:

Hi, what kind of document did you send to the AI?

— Reply to this email directly, view it on GitHub https://github.com/ParisNeo/lollms-webui/issues/496#issuecomment-1962300671, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEF52KO4YK2Q3BS7RSX7J73YVGR3XAVCNFSM6AAAAABDS2ANEKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTSNRSGMYDANRXGE . You are receiving this because you authored the thread.Message ID: @.***>