[GH-ISSUE #10004] Add full support for omni models #32318

Closed
opened 2026-04-22 13:27:55 -05:00 by GiteaMirror · 12 comments
Owner

Originally created by @flexiworld on GitHub (Mar 26, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10004

I want to use Qwen2.5-Omni-7B for images, audio, and video. Not only for text.

Originally created by @flexiworld on GitHub (Mar 26, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10004 I want to use Qwen2.5-Omni-7B for images, audio, and video. Not only for text.
GiteaMirror added the model label 2026-04-22 13:27:55 -05:00
Author
Owner

@dickens88 commented on GitHub (Mar 27, 2025):

I didnt see Qwen2.5-Omni-7B in the ollama model market yet

<!-- gh-comment-id:2759026268 --> @dickens88 commented on GitHub (Mar 27, 2025): I didnt see Qwen2.5-Omni-7B in the ollama model market yet
Author
Owner

@Himanshu8881212 commented on GitHub (Mar 28, 2025):

+1

<!-- gh-comment-id:2761863062 --> @Himanshu8881212 commented on GitHub (Mar 28, 2025): +1
Author
Owner

@hdcola commented on GitHub (Mar 29, 2025):

I think there will be more and more multi-modal models. I hope that ollama can support and begin to be compatible with mainstream WebRTC and WebSocket real-time APIs.

<!-- gh-comment-id:2763068704 --> @hdcola commented on GitHub (Mar 29, 2025): I think there will be more and more multi-modal models. I hope that ollama can support and begin to be compatible with mainstream WebRTC and WebSocket real-time APIs.
Author
Owner

@zhaojigang commented on GitHub (Mar 31, 2025):

+1

<!-- gh-comment-id:2766455188 --> @zhaojigang commented on GitHub (Mar 31, 2025): +1
Author
Owner

@Ramachandra-2k96 commented on GitHub (Mar 31, 2025):

+1

<!-- gh-comment-id:2767163150 --> @Ramachandra-2k96 commented on GitHub (Mar 31, 2025): +1
Author
Owner

@olumolu commented on GitHub (Mar 31, 2025):

Ollama does not support multimodal stuff at the point of time but i think they will working on supporting this as phi4 multimodal model support is also pending.

<!-- gh-comment-id:2767363028 --> @olumolu commented on GitHub (Mar 31, 2025): Ollama does not support multimodal stuff at the point of time but i think they will working on supporting this as phi4 multimodal model support is also pending.
Author
Owner

@Feixu2015 commented on GitHub (Apr 6, 2025):

+1

<!-- gh-comment-id:2781457580 --> @Feixu2015 commented on GitHub (Apr 6, 2025): +1
Author
Owner

@olumolu commented on GitHub (Apr 6, 2025):

I didnt see Qwen2.5-Omni-7B in the ollama model market yet

https://huggingface.co/Qwen/Qwen2.5-Omni-7B

<!-- gh-comment-id:2781528893 --> @olumolu commented on GitHub (Apr 6, 2025): > I didnt see Qwen2.5-Omni-7B in the ollama model market yet https://huggingface.co/Qwen/Qwen2.5-Omni-7B
Author
Owner

@xszconfig commented on GitHub (Apr 7, 2025):

@rick-github @olumolu Hi there!
Is there any update or roadmap that can be shared now?
Thank you for all the effort put into supporting Qwen 2.5 Omni.

<!-- gh-comment-id:2782444894 --> @xszconfig commented on GitHub (Apr 7, 2025): @rick-github @olumolu Hi there! Is there any update or roadmap that can be shared now? Thank you for all the effort put into supporting Qwen 2.5 Omni.
Author
Owner

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

Patrick and Bruce are actively working on qwen-2.5-vl and qwen-2.5-omni (although only conversion atm) support. Since this is likely to be folded in to the same release for qwen2.5-vl support, I'm going to mark this as a duplicate of #6564, please follow along there.

<!-- gh-comment-id:2783449291 --> @rick-github commented on GitHub (Apr 7, 2025): Patrick and Bruce are actively working on [qwen-2.5-vl](https://github.com/ollama/ollama/compare/main...brucemacd/qwen25vl) and [qwen-2.5-omni](https://github.com/ollama/ollama/compare/main...qwen25omni) (although only conversion atm) support. Since this is likely to be folded in to the same release for qwen2.5-vl support, I'm going to mark this as a duplicate of #6564, please follow along there.
Author
Owner

@tomasmcm commented on GitHub (May 16, 2025):

Issue https://github.com/ollama/ollama/issues/6564 was closed due to the recent release for multimodal models in Ollama, but it only supports the Qwen-2.5-VL and only images. The omni model adds audio in and speech out, so this issue is not a duplicate and is still required. Can you reopen it?

<!-- gh-comment-id:2886020653 --> @tomasmcm commented on GitHub (May 16, 2025): Issue https://github.com/ollama/ollama/issues/6564 was closed due to the recent release for multimodal models in Ollama, but it only supports the Qwen-2.5-VL and only images. The omni model adds audio in and speech out, so this issue is not a duplicate and is still required. Can you reopen it?
Author
Owner

@3unnycheung commented on GitHub (Jul 29, 2025):

any progress?

<!-- gh-comment-id:3131955916 --> @3unnycheung commented on GitHub (Jul 29, 2025): any progress?
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#32318