[GH-ISSUE #10392] Gemma3 add support for do_pan_and_scan #53340

Closed
opened 2026-04-29 02:40:30 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @wbste on GitHub (Apr 24, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10392

Reading the docs and looking at other inference projects, seems ollama doesn't implement do_pan_and_scan. Setting it totrue would greatly increase non square, high resolution image support if I'm reading it correctly.

https://huggingface.co/blog/gemma3#multimodality

Originally created by @wbste on GitHub (Apr 24, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10392 Reading the docs and looking at other inference projects, seems ollama doesn't implement ` do_pan_and_scan`. Setting it to`true` would greatly increase non square, high resolution image support if I'm reading it correctly. https://huggingface.co/blog/gemma3#multimodality
GiteaMirror added the feature request label 2026-04-29 02:40:30 -05:00
Author
Owner

@fakerms commented on GitHub (Oct 29, 2025):

Currently what preprocess will be done by ollama if image size is not the same with that encoder expected? crop? zoom?

<!-- gh-comment-id:3459320731 --> @fakerms commented on GitHub (Oct 29, 2025): Currently what preprocess will be done by ollama if image size is not the same with that encoder expected? crop? zoom?
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#53340