Replies: 1 comment
-
|
Have you also changed the embedding model? The default embedding model is set to OpenAI, which is likely why you’re getting an error when trying to use Vault Chat. As for the issue of “giving nonsense answers,” it seems to be a problem with the small model. Small models don’t handle long prompts well and tend to generate nonsensical responses when there’s a long context. This is a known issue: |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
When I try to use Chat, nothing works. When I try to use Vault chat to search all my files, I get an error message (see screenshot). The error message is asking for Open AI API key, but I am using Ollama as a provider.
When I try to ask a simple question attaching a single file as context, the Chat will reference another file entirely and its answer will be completely irrelevant. (see screenshot - where I am attaching a specific note and the Smart Composer chat references an entirely different note and gives a random answer).
Please help. :) Thank you.
Beta Was this translation helpful? Give feedback.
All reactions