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
Knowledge is a section in Open WebUI where you can store structured information that the system can refer to during your interactions.
It’s like a memory system for Open WebUI that allows it to pull from saved data, making responses more personalized and contextually aware.
You can use Knowledge directly in your chats with Open WebUI to access the stored data whenever you need it.
I am unsure how this would functionally work however. Would it call a say, DB query based on user input and/or would the system "surmise" what keywords to search for in the database based on the user input? Or is it just added as a "system prompt" (which again, tends not to work very well if it is long).
More details
Currently, "memories" can somewhat be emulated with Custom System Instructions in parameters for each model and template. This is pretty clunky however, not to mention I have to mix both instructions for the model and what it should remember about me in the same textbox.
Which components are impacted by your request?
General
Pictures
Code of Conduct
I agree to follow this project's Code of Conduct
The text was updated successfully, but these errors were encountered:
What features would you like to see added?
Something akin to ChatGPT memories, or how OpenWebUI handles memories/knowledge might work?
https://docs.openwebui.com/features/workspace/knowledge/
I am unsure how this would functionally work however. Would it call a say, DB query based on user input and/or would the system "surmise" what keywords to search for in the database based on the user input? Or is it just added as a "system prompt" (which again, tends not to work very well if it is long).
More details
Currently, "memories" can somewhat be emulated with Custom System Instructions in parameters for each model and template. This is pretty clunky however, not to mention I have to mix both instructions for the model and what it should remember about me in the same textbox.
Which components are impacted by your request?
General
Pictures
Code of Conduct
The text was updated successfully, but these errors were encountered: