[GH-ISSUE #8419] Does ollama support video as input? #5408

Closed
opened 2026-04-12 16:38:49 -05:00 by GiteaMirror · 4 comments
Owner

Originally created by @papandadj on GitHub (Jan 14, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/8419

For example, models like minicpm already support video. Can these models directly take video as input?

Originally created by @papandadj on GitHub (Jan 14, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/8419 For example, models like minicpm already support video. Can these models directly take video as input?
GiteaMirror added the feature request label 2026-04-12 16:38:49 -05:00
Author
Owner

@rick-github commented on GitHub (Jan 14, 2025):

Not currently. The ollama team are considering supporting different types of models, but that is some time off.

<!-- gh-comment-id:2589598719 --> @rick-github commented on GitHub (Jan 14, 2025): Not currently. The ollama team are considering supporting different types of models, but that is some time off.
Author
Owner

@papandadj commented on GitHub (Jan 15, 2025):

Thank you for considering additional model support! Just a friendly suggestion - would it be possible to prioritize video and video streaming support first? For most users, the ability to handle video inputs would be a significant capability expansion that could enable many more real-world applications. While model variety is valuable, video support seems more fundamental for expanding what users can actually do with the system. Just sharing my thoughts on development priorities! 😊

<!-- gh-comment-id:2591428053 --> @papandadj commented on GitHub (Jan 15, 2025): Thank you for considering additional model support! Just a friendly suggestion - would it be possible to prioritize video and video streaming support first? For most users, the ability to handle video inputs would be a significant capability expansion that could enable many more real-world applications. While model variety is valuable, video support seems more fundamental for expanding what users can actually do with the system. Just sharing my thoughts on development priorities! 😊
Author
Owner

@rick-github commented on GitHub (Jan 28, 2025):

Thanks for your thoughts. The ollama backend is undergoing a lot of changes at the moment, with the goal of making it easier to add a variety of models.

<!-- gh-comment-id:2620071966 --> @rick-github commented on GitHub (Jan 28, 2025): Thanks for your thoughts. The ollama backend is undergoing a lot of changes at the moment, with the goal of making it easier to add a variety of models.
Author
Owner

@Sirfrummel commented on GitHub (Mar 30, 2025):

Why is this closed? This sounds like a really cool feature to have. In the Gemma3 release announcement, they said it supported video. However, I haven't seen much mention about it. Would at least be cool to keep open and on the radar.

<!-- gh-comment-id:2764711454 --> @Sirfrummel commented on GitHub (Mar 30, 2025): Why is this closed? This sounds like a really cool feature to have. In the Gemma3 release announcement, they said it supported video. However, I haven't seen much mention about it. Would at least be cool to keep open and on the radar.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#5408