We recently sat down with three Google DeepMind team members to hear about their shared vision for our Omni model, what exactly makes it so helpful for creators, and why they focused the model on video generation to start. Plus, see the team put Omni in action with a... few visual experiments ↓show more

@GoogleDeepMind I love worthless AI slop thank you Google ❤️

@GoogleDeepMind Interesting here is that aach time a scene moves to another model, some of the brief may get lost, right. So, I’d keep editing the same scene in Omni for a while and see how much of the original direction it holds onto.

@GoogleDeepMind Chris is giving kinda chic energy

@GoogleDeepMind This is really a good news for content creator.

@GoogleDeepMind @Google curious to see how Omni handles different video styles. can it keep up with fast-paced editing or is it more for laid-back content?

@GoogleDeepMind Video first makes sense for one boring reason: temporal consistency is where models fail loudest. One bad frame and everyone sees the bug. Good stress test.

@GoogleDeepMind Video is the right stress test one bad frame is a bug users can actually see

@GoogleDeepMind The interesting part isn’t just that Omni can generate video. It’s that video might be the best place to prove whether a truly “omni” model actually understands what creators are trying to make. Can’t wait to see what people break with it.

@GoogleDeepMind Is this true Google?

@GoogleDeepMind Amazing! 👏 Loved hearing the vision behind Omni! The focus on video generation for creators is truly the future. The team at Google DeepMind is doing incredible work! Can't wait to see what creators will build with it! #GoogleOmni

@GoogleDeepMind Is he from India 👀

@GoogleDeepMind Omni's video hype? Like, why are we still waiting for actual results, lol.

@GoogleDeepMind PUT GOOGLE CHROME ON MOBILE BACK TO HOW IT WAS. NOBODY ASKED FOR THIS.

@GoogleDeepMind Starting with video inverts the text-first paradigm. The real experiment: does spatial-temporal grounding build different world representations than text training? If so, it changes what "understanding" means downstream.

@GoogleDeepMind Rất thú vị khi nghe trực tiếp từ đội ngũ về định hướng Omni và lý do họ ưu tiên khả năng tạo video cho nhà sáng tạo

@GoogleDeepMind Great behind-the-scenes look at the vision and creative potential of Omni.

@GoogleDeepMind

@GoogleDeepMind Video generation gets much more useful when it fits into the creator's workflow. The next interesting step is seeing how models like Omni connect ideation, iteration, and production in one continuous loop.

Google DeepMind is betting that AI creators need more than another video model. Omni is being built around a broader vision for visual experimentation and creation. the team recently showed how they're thinking about that through a series of demos. AI video is becoming a much bigger interface for creativity

@GoogleDeepMind

@GoogleDeepMind Help me to reactivate my Gmail ID

@GoogleDeepMind

@GoogleDeepMind So cool hearing creators get some love here. But I hope Omni doesn't end up making everything feel samey. Anyone else hoping there's room for weird, glitchy outputs too, not just polished clips?

@GoogleDeepMind Starting with video makes sense since creators feel visual consistency issues first

@GoogleDeepMind DeepMind out here teaching AI how to generate video while my tailor still takes three weeks just to get a collar roll right. Priorities are completely upside down.

@GoogleDeepMind Holy shit. That's terrible PR. If this vid got views or Moist Critikal watched it... This will start a global anti-AI trend, concluding with data center terrorism or anti-AI terrorism. Please don't talk about how Omni can make stories, the rabid artists are gonna make sure y'all

@GoogleDeepMind Amazing insight Omni is truly empowering creators everywhere!

@GoogleDeepMind Gemini Omni Flash has completely changed the way I create videos. 🚀 Its Avatar feature has made video creation faster and more accessible, while its powerful editing capabilities make it incredibly easy to add motion graphics, refine scenes, and simplify the entire workflow❤️

@GoogleDeepMind

@GoogleDeepMind