Creative producer specialising in AI production. Google DeepMind.
London
- Tweets 171
- Following 122
- Followers 1,759
- Likes 579
🎉 Today, we’re making Gemini 2.5 Deep Think available to Google AI Ultra subscribers in the @GeminiApp
Gemini 2.5 Deep Think is a variation of the model that achieved the gold-medal standard at this year’s International Mathematical Olympiad (IMO). Deep Think uses parallel thinking and reinforcement learning techniques to deliver more detailed and thoughtful responses.
To demonstrate this, we prompted the model to generate a voxel art image of a pagoda using a single prompt. As you can see, compared to 2.5 Pro, Deep Think adds more and more detail to every aspect of the scene. From the architecture of the pagoda, to the fidelity of the trees, to the diversity of the ground plane, every aspect of the prompt gets handled more deeply.
2
18
1,414
This is the real me, but the background is courtesy of Gemini 4 Argon.
I showed Argon the two green-screen stills below and asked it to 1) build me a room in Blender, 2) recreate the camera angles, 3) use Nano Banana to 'skin' the Blender background to match the foreground, and 4) key it back into the original video.
The result: an accurate background across both angles.
Gemini 4 Argon is now globally available.
Built to solve your hardest problems from start to finish, the model brings frontier capabilities across coding, enterprise workflows, and cyber defense.
42
49
781
45,568
Gemini 3.7 Flash could make a decent DOOM (1992) clone from a single prompt. Gemini 3.8 Flash can make a decent GoldenEye (1997) clone.
That's a five year jump in two weeks.
10
14
217
20,852
I used Gemini 3.7 Flash in @antigravity to generate an Omni video in Google Sheets... (1/3)
44
48
930
259,808
The secret to beautiful games isn't just good coding—you also need great art assets. I made this demo that combines Gemini 3.7 Flash with Nano Banana to gorgeous, fun-to-play games, quickly and affordably, with a single prompt. 🪄
2
1
27
3,832
Making amazing motion text is easy. Thinking of clever things to write isn't!
I vibe coded this AIS applet that combines Nano Banana Pro's text and image gen with Veo 3.1's video gen capabilites. You can try it here: ai.studio/apps/bundled/type_…
3
576
Have you ever been to a museum where you press a button at an exhibit and listen to a historical reenactment? I vibe coded an app that enables you to do that anywhere in the world, for any time in history! Made with Gemini 3. Try it here: ai.studio/apps/bundled/voice…
3
545
Try this prompt in Gemini 3 (with a photo):
"You are Sherlock Holmes, known for your amazing powers of observation and logical deduction. What deductions can you make from this image that the ordinary person would miss? Astound me."
It's crazily accurate! 🔎
1
17
1,150
If you make a particualrly cool build, you can share it by copying the JSON data and sending it to a friend. They can paste it into their app and see what you built.
1
1
145
I used Gemini 3 Pro to build this 'voxel toy box' app.
The app will generate a beautiful sculpture for you using voxels (sort of like 3D pixels).
Here's where it gets interesting: you can then ask Gemini to smash the sculpture and use the same blocks to build something else.
1
257
I used this Nano Banana app to create this water colour of one of my favourite places in London, Brompton Cemetery. Also one of my cat's favourite places in London!
aistudio.google.com/apps/bun…
1
5
300
Asking kids to imagine the future using AI | Morning Star AI piped.video/watch?v=HgTn7ZHP…
145
Now and Then | John Lennon deepfake piped.video/3R_mHMG8TaU?feature…
138
Gormenghast | AI-generated video sketches (2023) piped.video/ApHKkH9nvQs?feature…
126
Am I Even Speaking To A Real Person? | RUNWAY GEN:48 submission | George Kenwright piped.video/5Kd-NW0KFIw?feature…
118