Fun test of Gemini Omni's video editing and location knowledge ✨
I uploaded a video riding in a Waymo in Menlo Park.
Then I asked Omni to re-shoot the video in different locations based on screenshots from Google Maps.
It's seamless, as you can see from the transitions 👇
Text fidelity was a big focus for us.
Video is full of micro-text—phone screens, street signs, background books. If an AI model can't handle those details, it breaks the immersion. Fixing it in post is hard and painful.
Getting text generation right is foundational to getting video right. Also love the @fofrAI shout out!
When we began testing #GeminiOmni I wanted to see the accuracy of the edits. So I made a tool directly in @FlowbyGoogle. DIFF TOOL allows you to compare images/videos in realtime or export videos in sync. Try it out or remix it to make it your own.
https://t.co/SAh71v00Ff
6 months ago I packed up my life and moved back to the Bay Area. Now you get to see why 🥳
Gemini Omni
It understands the world. It creates from any input. You edit with conversation.
Gemini Omni Flash is rolling out starting today.
Here’s where you can find it:
🔹 Today: Google AI Plus, Pro and Ultra subscribers globally in the @GeminiApp and @FlowbyGoogle .
🔹Rolling out starting this week, for no cost: @YouTube Shorts and the YouTube Create app.
🔹Coming weeks: developers and enterprise customers via APIs.
#GoogleIO
Nano Banana 2 delivers some serious gains 💪 😀
+71pts: 3D Imaging & Modeling
+60pts: Text Rendering
+57pts: Product, Branding & Commercial Design
+52pts: Cartoon
Huge congratulations to @nainar92@heyitsbeaaaaaaa
🔍Let’s take a closer look at Text-to-Image subcategories and compare how the top 3 perform: Nano Banana 2 vs. GPT-Image-1.5 vs. Nano Banana Pro.
Nano Banana 2 (Gemini-3.1-Flash-Image-Preview) largest gains over Nano Banana Pro (Gemini-3-Pro-Image-Preview-2k):
- +71pts: 3D Imaging & Modeling
- +60pts: Text Rendering
- +57pts: Product, Branding & Commercial Design
- +52pts: Cartoon
Where @OpenAI’s GPT-Image-1.5 still leads over Nano Banana Pro (Gemini-3-Pro-Image-Preview-2k):
- +16pts: Portraits
- +14pts: Cartoon
- +13pts: Product, Branding & Commercial Design
Congrats again to @GoogleDeepMind for this leap forward today!
Josh might be the most humble exec at Google. I think that's where the magic comes from. Such a good read and look into his process: https://t.co/PNpvIpsUQx
Nano Banana's world knowledge is 🤯
A little late to this trend but its amazing that it actually generated a little hand monument!
context: https://t.co/YR7KUGzaKX
After almost 2 years of building the video gen API business for Google Cloud, thrilled to share that I joined @GoogleDeepMind
Ill be working on the next generation of video generation models with the amazing team that brought you Veo 3 and Nano Banana.
@nbrichtova@kinwong@nainar92@heyitsbeaaaaaaa