@poolio

ex-google brain and deepmind. phd in neural nonsense from stanford.

 San Francisco, CA
poolio.github.io
Joined March 2008
  • Tweets3.1K
  • Following1.5K
  • Followers23.4K
193Media
Ben Poole retweeted
SpAItial AI
@SpAItial_AIOct 6
We're hiring at SpAItial 🚀🚀🚀 Want to build AI that generates 3D worlds and understands their physics? 15 roles in London & Munich: 🧠 RS: World Models · 3D Diffusion · 3D Recon · Robot Learning 🛠️ RE: World Models · 3D Recon · Graphics ⚡ ML Infra · Inference Join us 👇
11
16
4
88
22,186
Congratulations to our first #NobelPrize laureate of 2026 🎉 Early this morning Karl Deisseroth found out he was awarded this year's Nobel Prize in Physiology or Medicine. He seems very happy with the news!
Karl Deisseroth on the phone looking happy

ALT Karl Deisseroth on the phone looking happy

139
963
124
12,155
569,030
Ben Poole retweeted
Alex Albert
@alexalbert__Sep 22
I've been on a Blender kick with Opus 5.5. Its better 3D modeling and vision mean you can build an entire world from a single prompt. Historically accurate San Francisco Market street in 1906, pre-earthquake
97
50
33
1,163
165,319
Ben Poole retweeted
Alex Albert
@alexalbert__Sep 22
I just realized Opus 5.5 can use Blender to make claymations with just a single prompt in claude dot ai now
107
107
35
2,640
269,832
Ben Poole retweeted
Sander Dieleman
@sedielemSep 14
We are hiring research scientists and engineers🧑‍🔬 on the Gemini Omni team! Both in the US (Mountain View / SF) and Europe (London / Zürich). Apply here: google.com/about/careers/app…
11
67
10
777
155,969
Ben Poole retweeted
Keunhong Park
@KeunhongPSep 1
Atlas is live! A spatial foundation model, trained in-house from scratch. After shipping RTFM last year, I was convinced a single camera-conditioned model could unify most spatial generation tasks. Atlas is that bet at scale. The core idea👇
World Labs
@theworldlabsSep 1
Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.
4
15
1
179
61,920
Ben Poole retweeted
Robin Rombach
@robrombachAug 4
Two years later our first video model - FLUX 3 Video - is finally released🥹. This is such a strong, fun and versatile model made by the best team in the world @bfl_ai. We will add further variants and generation from references in the near future. I am also super excited about the upcoming open weight, action prediction and image variants of FLUX 3.
37
51
6
496
26,314
Ben Poole retweeted
Shaun Singh
@shaunxsinghJul 16
Fortunate to work with this team every day, ACT-2 is a big leap. Working on real Solves is a blast. Next Stop: Recursive Solve Improvement
Tony Zhao
@tonyzzhaoJul 16
Introducing ACT-2 Preview The first robotics model to unify broad generalization with high reliability. A single fine-tuning example can teach Memo a new behavior that generalizes. Zero shot, real unseen homes, 99% success rate.
6
5
1
48
3,182
Ben Poole retweeted
Oliver Wang@oliver_wang2Jun 30
1/2 Introducing Nano Banana 2 Lite! We are working hard at the next frontier model, but in the meantime enjoy this fast and cheap Nano Banana (Pico Banana?). It returns images in only a few seconds, and with most of the smarts of the NB2 model! deepmind.google/models/gemin…

Nano Banana 2 Lite

Google DeepMind

deepmind.google
2
8
47
4,225
We’re shipping 2 major releases:
 🔘 Nano Banana 2 Lite: our fastest and cheapest Gemini Image model 🔘 Gemini Omni Flash: now available via the Gemini API and in @GoogleAIStudio to help developers generate and edit high-quality videos.
101
194
57
1,118
282,663
Huge congratulations to the Project Genie team on taking home the Cannes Lions Grand Prix for AI Craft! 🎉 Thank you @Cannes_Lions for recognizing how Genie pushes beyond conventional creative limitations to achieve outcomes unattainable without frontier AI… Another unexpected consequence of scaling world models :) Try it yourself here: labs.google/projectgenie
22
61
10
323
114,257
Ben Poole retweeted
Arena.ai
@arenaJun 11
Exciting news: Gemini Omni Flash is now #1 in the Video Arena (both Text-to-Video and Image-to-Video)! For Text-to-Video this is a massive +158 pt improvement over Veo 3.1 (1080p) and a large +61 pt lead over the next best model, Seedance 2.0. Congrats @GoogleDeepMind for this huge milestone!
Google DeepMind
@GoogleDeepMindMay 19
We’re dropping Gemini Omni: our first step towards a model that can create anything from anything - starting with video. It combines Gemini’s intelligence with our generative media systems - representing a leap forward in world understanding, multimodality, and editing 🧵
145
238
69
1,500
434,175
Ben Poole retweeted
Google Labs
@GoogleLabsJun 10
🌍 Project Genie access is expanding even more! Starting today, Google AI Ultra 5X subscribers (our latest tier!) globally can access Project Genie. Try it out here! labs.google/projectgenie

Project Genie

Project Genie is an experimental prototype that lets you create and explore infinitely diverse worlds.

labs.google
Google Labs
@GoogleLabsMay 20
We love seeing all of the worlds you’ve been creating with Genie. SO much so, that we’re excited to announce Project Genie is now fully available to all Google AI Ultra subscribers globally (18+).
33
118
14
748
937,910
Ben Poole retweeted
Sumith Kulal
@sumith1896Jun 6
Diffusion Circle happening today @CVPR #CVPR2026 Upper Lobby F at 3:30pm, come join us!
1
14
4
52
22,062
Ben Poole retweeted
Ideogram
@ideogram_aiJun 5
Ideogram 4.0 has four key components: (1) a frozen Qwen3-VL-8B text encoder (2) a 34-layer single-stream DiT (3) a flow-matching Euler sampler with asymmetric CFG (4) a frozen FLUX.2 VAE We thank the open-source community and give back by releasing this with open weights.
2
6
1
63
8,225
Try out our image to video feature in Gemini Omni Flash! Interested to see what people are able to do.
Google Flow
@FlowbyGoogleJun 4
Bring your images to life ⚡️ Upload your picture as a first frame and add a prompt to generate your own unique video with Gemini Omni Flash. Tag us and share your results!
1
3
1
12
3,579
Ben Poole retweeted
Robin Rombach
@robrombachJun 5
Diffusion Circle with forest vibes! 🌲 Since the original Diffusion Circle organizer @sedielem cannot make it to CVPR in person this year, we coordinated with him and are happy to host the Diffusion Circle this time. Stop by if you are interested in flow maps, multimodal diffusion models, variational flow matching, action prediction, and all the other hot denoising stuff. @sumith1896 and @dustin_podell will coordinate. Meeting point: Upper Lobby F / Exhibit Hall F Entrance; see attached pics. Happening Saturday, June 6, 3:30pm!
8
15
3
106
16,422
Ben Poole retweeted
Daniel Yamins
@dyaminsJun 4
The PSI blog will be a good place to look for world model updates: neuroailab.github.io/psi-web…
2
3
20
2,987
For the past years my research focus was on unifying models and training paradigms across modalities. Today I'm excited that we're releasing our latest model aligned with this theme: Gemma 4 12B, a dense encoder-free model which processes raw text, image, and audio inputs! 1/
28
123
19
1,112
111,611
Ben Poole retweeted
Tripo
@tripoaiJun 1
Introducing Project Eden, a world model research preview from @VASTAIResearch Project Eden is a persistent, multiplayer world model that fundamentally breaks from existing paradigms by decoupling the underlying world state from visual rendering. Instead of treating the world as a sequence of transient frames, Eden treats it as a structured, evolving environment that runs continuously, can be modified by user actions, and can be consistently observed from any viewpoint.
92
143
33
852
3,413,081
Load more