Impressive new Gemini ‘Omni’ video model leaked ahead of Google I/O

If you’ve been following the AI ​​field lately, you know that video generation is the current frontier. While OpenAI appears to be putting the brakes on the Sora model, Google has repeatedly stated that “video is here to stay.” We’ve known about the Veo model for a while, but according to a new report: 9to5 google, It looks like there’s a whole new iteration called gemini omni It’s already starting to show up in the wild for some users.

What is Gemini Omni?

According to the leak, at least one Gemini user was prompted to “Create with Gemini Omni.” Google describes the new video generation model as a way to remix videos, edit them directly in chat, and try out ready-made templates.

It’s not entirely clear how Omni fits into the broader context of Gemini and existing Veo models, but metadata suggests that Gemini Omni may actually be an extension or evolution of the Veo foundation. Regardless of the naming convention, the early results are very impressive.

Pass the spaghetti test

If you remember the early days of AI video, you probably remember the horrifying viral clip of Will Smith eating spaghetti. It became an unofficial benchmark. bad It could be an AI video. Well, one of the leaked Omni demos specifically used a prompt for two men eating spaghetti in an upscale restaurant, and the results were incredibly realistic.

Featured videos

Another demonstration saw a professor writing out mathematical proofs of trigonometric identities on a traditional blackboard as he explained the steps. There are still some obvious AI “instructions” in the final output, but the model handles text generation and physical movement very well.

Know usage restrictions

Of course, generating ultra-realistic videos in chat windows requires a ton of computing power, and Google seems to be fully aware of that. Users who found these Omni prompts also noticed a new “Usage” tab in their accounts. I managed to understand it by only generating these two video prompts 86% of daily usage With the AI ​​Pro plan.

Google’s intentions to add more explicit usage restrictions to Gemini were recently revealed, and resource-intensive models like Omni may be exactly why. With Google I/O 2026 just around the corner, I’m confident Gemini Omni will take the stage in earnest. We’ll be watching closely to see how this plays out and what it means for the future of video production at Gemini.