François Chollet
@fchollet
Co-founder @ndea. Co-founder @arcprize. Creator of Keras and ARC-AGI. Author of 'Deep Learning with Python'.
Big news: we just released Keras 3.0! ▶ Run Keras on top of JAX, TensorFlow, and PyTorch ▶ Train faster with XLA compilation ▶ Unlock training runs with any number of devices & hosts via the new Keras distribution API It's live on PyPI now! 🚀
Please note, we're not able to reproduce the 41.8% ARC-AGI-1 score claimed by the latest Qwen 3 release -- neither on the public eval set nor on the semi-private set. The numbers we're seeing are in line with other recent base models. In general, only rely on scores verified by…
Check out zea, a flexible toolbox for cognitive ultrasound imaging built on Keras 3 ✔️ Signal processing & image reconstruction ✔️Pretrained models ✔️Data loading tools ✔️Multi-backend support via Keras 3 (you can use JAX, TF, and PyTorch)
New ARC Prize 2025 High Score 17.6% by Giotto. ai (@podesta_aldo)
A hallmark of human intelligence is the capacity for rapid adaptation, solving new problems quickly under novel and unfamiliar conditions. How can we build machines to do so? In our new preprint, we propose that any general intelligence system must have an adaptive world model,…
Officially validated IMO gold medal, purely via search in token space, achieved in 4.5 hrs (unclear at what compute cost). The solutions read nicely as well deepmind.google/discover/blog/…
With ARC-AGI-3 we now have a new tool In addition in measuring efficiency via cost (like we do with ARC-AGI-1/2) We now can measure efficiency via “action count” How many moves does it take you complete a game? We run random agent against all games, so we’ll have a floor to…
Intelligence isn't a collection of skills. It's the efficiency with which you acquire and deploy new skills. It's an efficiency ratio. And that's why benchmark scores can be very misleading about the actual intelligence of AI systems.
Intelligence isn't a collection of skills. It's the efficiency with which you acquire and deploy new skills. It's an efficiency ratio. And that's why benchmark scores can be very misleading about the actual intelligence of AI systems.
North to South, we're covering more ground in Austin starting today. Safe and fully autonomous Waymo rides, only on @Uber. See you on the road, ATX.
Team 'The Architects' just reached the top of the leaderboard!
New ARC Prize 2025 High Score 15.8% by The ARChitects x.com/arcprize/statu…
The goal of education should not be to achieve a certain population average on standardized tests, it should be to unlock the highest total individual potential out of students