Search: “francois-chollet--x”

Cross-agent feedback loops are incredibly effective -- for a reason. Check out what @leon2mcp and team at @Bloome_im are building in this space: http://bloome.im Bloome lets you pull Claude, ChatGPT,

Quick reminder of what's ok vs not ok with harnesses used for playing ARC-AGI-3: 1. Not okay: harnesses that were custom-made to solve the benchmark or that contain knowledge about the benchmark forma

We re-tested GPT-5.6 Luna from @OpenAI on ARC-AGI (Verified) following its recent 80% price reduction: - ARC-AGI-2: 59.6%, 0.18/task - ARC-AGI-1: 90.7%, 0.07/task The new results match Luna's original

GPT-6 Astra by @OpenAI achieves SOTA on ARC-AGI: - Astra scores 63% on ARC-AGI-3, 99% via a new provider adapter harness - It surpasses human performance on 96% of ARC-AGI-3 levels - It builds the mos

> One could even define intelligence as the efficiency with which one converts experience into competence; by this definition they lag very far behind us Yes, one could define intelligence in this way

I think 'AI welfare' is morally obscene (given all the human suffering in the world), and will add to the growing anti-AI backlash. AI is a technology that humans created that should be used for our b

GPT-6 Sol from @OpenAI on ARC-AGI (Verified): - ARC-AGI-3: 4.6%, 5.6K (standard harness), 23.0%, 8.7K (provider adapter harness) - ARC-AGI-2: 89.6%, 0.44/task - ARC-AGI-1: 95.5%, 0.14/task Sol reaches

RL is a bit of a double edged sword: in known territory performance increases, but in unknown territory the model tends to hallucinate that it is performing a completely different task it was trained

Gemini Gemini 3.5 Flash ARC-AGI (Verified) ARC-AGI-2: - High: 72.1%, 0.85 - Minimal: 8.9%, 0.11 ARC-AGI-1: - High: 92.5%, 0.42 - Minimal: 48.8%, 0.06 Gemini 3.5 Flash is on par with GPT-5.5 (Medium) o

Cross-agent feedback loops are incredibly effective -- for a reason. Check out what @leon2mcp and team at @Bloome_im are building in this space: http://bloome.im Bloome lets you pull Claude, ChatGPT,

We re-tested GPT-5.6 Luna from @OpenAI on ARC-AGI (Verified) following its recent 80% price reduction: - ARC-AGI-2: 59.6%, 0.18/task - ARC-AGI-1: 90.7%, 0.07/task The new results match Luna's original

François Chollet commented that claims the demise of Google were greatly exaggerated, cautioning against undervaluation. According to a Polymarket report, Sergey Brin is expected to take direct oversi

Introducing Gemini 3.8 Flash, another jump in Gemini's agentic + coding capabilities, and our 3rd updated Flash model in only 6 weeks... This model has been a ton of fun to work with, excited to see w

Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon! It shows frontier performance in complex workflows, cyber defen

Announcing ARC Prize Research Summit 2026 Research Speaker Kevin Murphy @sirbayes is a research scientist at Google DeepMind and author of the “Probabilistic Machine Learning” textbooks, developing AI

The new model from Meta is already looking like a disappointment: overoptimized for public benchmark numbers at the detriment of everything else. Knowing how to evaluate models in a way that correlate

Francois Chollet presents a framework for understanding different approaches to token usage in AI models: casual users maximize tokens for flexibility, competitive users minimize tokens for efficiency

Coding isn't yet another application domain -- it's the meta-skill required for AI to automatically develop its own training material, via symbolic world models. That's how the RSI loop actually kicks

Side note: when we released ARC-AGI-3 in March, and frontier models scored <1% on it, a few Singularitarian poasters took it as a personal insult, and got very worked up about it. They argued the benc

What is a way to test human-level generality in artificial intelligence? 'The meta-benchmark of being able to pass ARC-AGI-(n+1) immediately upon release' @Jabaluck @DAcemogluMIT The meta-benchmark of

There's no doubt that the world can consume tokens as fast as they're produced, even in the most maximalist infrastructure buildup scenarios imaginable. That's not the question. The question is whethe

Outstanding cost-to-performance from DeepSeek GPT-5.6 Luna (Max) performance for a 1/4th of the cost on ARC-AGI DeepSeek V4 Flash from @deepseek_ai on ARC-AGI (Verified): - ARC-AGI-2: 61.4%, 0.04/task

The power of JAX Introducing gyaradax 🐉: A JAX solver for local flux-tube gyrokinetics with custom CUDA kernels for acceleration. This entire code was vibecoded by @ggalletti_ and me in a month. Valid

This is very nice work from NVIDIA. Like all high-performing approaches on ARC-AGI-3, it uses deep learning-guided on-the-fly synthesis of symbolic world models, i.e. navigating the world by generatin