DeepMind during the day, modular synth guy at night: nerodiseppia.ch

Zürich, Schweiz
Glad we launched => aistudio.google.com
75
276
4,525
459,679
It seems like a lovely day in London. Use the prompt below on Nano Banana Pro to make cute images of a location with live weather conditions. Make sure you have search grounding enabled to get the current weather.
1
12
694
🤯🤯
Google’s Nano Banana Pro is by far the best image generation AI out there. I gave it a picture of a question and it solved it correctly in my actual handwriting. Students are going to love this. 😂
3
742
Prompt: "Make this into professional digital art. zoom out a bit and show more. Make it look engaging"
1
86
not sure what happened here lol
76
Fabian Mentzer retweeted
Replying to @mentzer_f
1
2
243
Check out these amazing examples from the 🐐 AS himself ⬇️⬇️
We are excited to release Nano Banana Pro today! We have been working hard for the past few months trying to improve the model and now can’t wait for everyone to play with it! The model should feel smarter for both t2i and editing and also comes with new capabilities.. (1/n)
1
4
566
Fabian Mentzer retweeted
None of these cool things would be happening without @phillip_lippe What Phillip pulled off, especially on thinking and image generation, is beyond impressive 🫡
Since I joined GDM in January, we've been working on making our image generation models smarter. From 2.0 Flash, Imagen 4, Nano Banana, and now Nano Banana Pro, it's been quite a ride with immense progress. And there's no sign of slowing down 🚀🍌
3
1
109
18,174
Very proud to have contributed to this amazing model!!! 🍌🍌 →→ check it out → →deepmind.google/models/gemin…
4
32
3,209
Fabian Mentzer retweeted
Replying to @ErikaCoaches
Confirmed
5
5
219
28,227
Fabian Mentzer retweeted
Nano Banana Pro NAILED my burguer test It's the first model to do this perfectly
79
34
2,063
189,613
Fabian Mentzer retweeted
Nano Banana Pro is wild. Here’s my favorite use case so far: take papers or really long articles and turn them into a detailed whiteboard photo. It’s basically the greatest compression algorithm in human history.
141
594
6,689
1,340,717
🍌🍌 👀
Currently in Japan 🍌 🍌
6
678
🍌🍌
2
2
16
1,775
🤯🤯🤯
One nice thing you can do with an interactive world model, look down and see your footwear ... and if the model understands what puddles are. Genie 3 creation.
2
437
Fabian Mentzer retweeted
Yesterday we announced Genie 3. One feature of the model that's especially fun to play with is starting worlds from existing videos. Here's a drone shot generated by Veo 3, with me taking control mid-flight.
What if you could not only watch a generated video, but explore it too? 🌐 Genie 3 is our groundbreaking world model that creates interactive, playable environments from a single text prompt. From photorealistic landscapes to fantasy realms, the possibilities are endless. 🧵
92
213
2,458
703,716
Fabian Mentzer retweeted
Incredible evolution of "Neural Video Games": from GQN (2018) to Genie3 (2025). The future is exciting! deepmind.google/discover/blo…
54
319
2,947
260,452
Dope image gen models ⬇️
We updated our Imagen 4 models and Ultra is tied for #1 on the lmarena leaderboard! The models are available in Google AI Studio and the Gemini API - try them out and let us know what you think.
3
327
Fabian Mentzer retweeted
Google's new video generator Veo 3 looks insane
16
104
1,655
144,905
Fabian Mentzer retweeted
Gemini-2.5-Pro-preview-05-06 isn't only the best model for coding. its the best model for everything! across all tasks!
35
112
1,404
224,104
Fabian Mentzer retweeted
Gemini 2.5 Pro latest update is amazing at code! What I find even more amazing is combining this coding skill with Gemini's state-of-the-art MM video understanding capability 🎞️, here transcribing a YouTube video ✈️ into a p5.js animation. PS: guess which video is this :)
🚨Breaking: @GoogleDeepMind’s latest Gemini-2.5-Pro is now ranked #1 across all LMArena leaderboards 🏆 Highlights: - #1 in all text arenas (Coding, Style Control, Creative Writing, etc) - #1 on the Vision leaderboard with a ~70 pts lead! - #1 on WebDev Arena, surpassing Claude for the first time This is the first-ever sweep across text, vision, and WebDev by any model!🥇 Huge congrats to @GoogleDeepMind on this incredible breakthrough!
2
21
132
16,240