Skip to content
Discussion options

You must be logged in to vote

When I was testing and writing code for GenAI, I saw this occasionally with the llava model (and others) when I was low on GPU memory and/or inference time was very high.

Since the requests are getting to Ollama, this isn't anything we can improve on the Frigate side.

Replies: 2 comments 16 replies

Comment options

NickM-27
May 22, 2025
Collaborator Sponsor

You must be logged in to vote
1 reply
@illuzn
Comment options

Comment options

You must be logged in to vote
15 replies
@illuzn
Comment options

@hawkeye217
Comment options

@illuzn
Comment options

@hawkeye217
Comment options

Answer selected by illuzn
@illuzn
Comment options

@zolakt
Comment options

@hawkeye217
Comment options

@zolakt
Comment options

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment