[GH-ISSUE #8914] New Mistral-Small need more quantization options #31540

Closed
opened 2026-04-22 12:05:10 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @co-search on GitHub (Feb 7, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8914

Hi

The new Mistral-small:24B is a great model
But, the smallest quantization available is Q4 => 14GB
It will be very beneficial to have Q3 option to fit the VRAM of 12GB cards, this will make the model runs much quicker
Thanks

Originally created by @co-search on GitHub (Feb 7, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8914 Hi The new Mistral-small:24B is a great model But, the smallest quantization available is Q4 => 14GB It will be very beneficial to have Q3 option to fit the VRAM of 12GB cards, this will make the model runs much quicker Thanks
GiteaMirror added the model label 2026-04-22 12:05:10 -05:00
Author
Owner

@rick-github commented on GitHub (Feb 7, 2025):

https://github.com/ollama/ollama/issues/8752

<!-- gh-comment-id:2642289036 --> @rick-github commented on GitHub (Feb 7, 2025): https://github.com/ollama/ollama/issues/8752
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#31540