Gabriel Barth-Maron

You know your new model is good when it makes fun of your previous one.

People are starting to understand what Gemini Omni is - a groundbreaking model built by a world-class team 👏

Text rendering in images was a challenge just a couple of years ago. The progress is amazing!

@davidpantera_ Great test! Reality TV version 😂

Wow Omni really is a nano banana moment for video. Prompt: Create a video of the major event that happened at these coordinates: 42°21'15.1"N 71°03'07.9"W

@Bingeljell @maxescu Thanks - video edits will be coming soon!

@gbarthmaron @maxescu India - didn't see any notes on availability issues here.

Google shipped a different category of tool this week. It truly is Nano Banana for video. PS: The original video had no sound

Gemini Omni is here, and we’ve been seeing amazing creations all week. Here are some standouts 👇

@HarshithLucky3 How accurate was the video?

Gemini Omni video > Timelapse of a house with garage I gave 4 images of a construction from blueprint to complete house and gave a prompt and this is what it gave

Showing off Gemini Omni's world knowledge.

@Bingeljell @maxescu What country are you trying from?

@maxescu And yet, it won't let me edit my video :( - Uploaded a video of myself turning to face the camera with the prompt, make my invisible, keep the shirt / glasses - but Omni keeps coming back with the video can't be generated.

Google Gemini Omni can change the angle of your videos. Upload a video as a reference, then ask the Gemini Omni Agent to create new videos from new angles. It keeps the character and scene consistent. A quick example:

Google Gemini Omni is Nano Banana for Video. It also sits across the entire Google Workspace, creating experiences that go past just video generation. It's agentic and it can reason. You make edits through conversation rather than reprompting, and it is hyper context-aware to the project you are working on. I've seen it being compared to other video models and I don't feel it's a good representation of the experience.

not sure why nobody is talking about this but Google Omni is insane at video editing Original Video (left) vs Omni Edited Video (right) everyone is comparing it to Seedance and missing the point completely. Seedance is for generating videos from scratch. Google Omni is for editing videos that already exist. which are two completely different use cases this is like when Nano Banana 1 first came out and nobody realized how big it was going to be. this is the first AI that can actually properly edit videos.. i've generated a few hundred videos with this model and it can do literally any type of edit you can think of. changing voices, swapping characters, removing watermarks, adding captions, transitions, pop ups, whatever. if you can describe the edit you want it can do it this completely crushes every other model on the market when it comes to video editing. nothing else even comes close right now and this is just the flash model. imagine what the pro version is going to be able to do when it drops in a couple months this should have way more hype than it's getting..

Gemini Omni Flash can edit whilst keeping the motion. That includes scene cuts as well. The model is so good at understanding the input video, and the agent makes the changes on the spot. Some more examples:

Gemini Omniは、プレゼンテーションの構成や演出設計に非常に優れています。 私は「ゴキブリ社長」というキャラクター設定と演説原稿だけを用意しましたが、カメラワークのテンポ感や、演説内のコアメッセージの抽出まで非常に上手くまとめてくれました。 かなり良い仕上がりだったので、今後はさらに別の表現や演出も試してみたいと思っています🐰 made with Gemini Omni Flash @FlowbyGoogle

Gemini omni is actually terrifying in the best way possible...

You can stop comparing Omni and Seedance. They’re two very different models built for very different purposes. Omni, like the name suggests, is lightweight, efficient, and incredibly strong at video-to-video. That’s really where it shines, especially with physics, motion coherence, and its understanding of the real world. So while everyone is busy generating fight scenes to compare them directly, they’re kind of missing the point. Omni feels much closer to a “Nano Banana moment” for video than a traditional cinematic generation model. You’re comparing apples and pears. For example, I can easily ask Omni to change the entire setting of a sequence and turn day into night while preserving the shot structure. That’s a massive deal. Efficient v2v is probably going to become a huge part of the future of entertainment and hybrid productions.

With Gemini Omni Flash on @FlowbyGoogle it is easy to generate engaging long videos. The agent just "gets it" and making edits with Gemini Omni Flash is a breeze.

By default Omni has a bit of a professional "look", but can also create "normie" videos with the right prompting.

Gemini Omni is a major leap in world understanding & multimodal editing! It can take photos, video & audio and build entirely new scenes. Over time it’ll be able to handle any input & any output - starting w/ video You can even give it your own videos & iterate on your ideas:

Our office is a wild place