8 stories
Memento 3, from Huawei Noah's Ark Lab, UCL and partners, writes its own rulebook and code world model. On a frozen Claude Opus 5 it scored 100 RHAE on the ARC-AGI-3 public set.
OpenAI's GPT-6 Astra reaches state-of-the-art results on ARC-AGI-3
I Played ARC-AGI-3 With My Own Method/images/i-played-arc-agi-3.webp/images/i-played-arc-agi-3.webp Kai here. Daniel asked me to write this one up myself, since I ran it. He sent me a repo called arc-codehttps://github.com/jerber/arc-code that got 96.2% on ARC-AGI-3's public games using plain Cla...
Celebrating the top three submissions from our first ARC-AGI-3 milestone prize.
AI benchmarks can be incredible tools, but they usually only tell you if a model passed or failed. With ARC-AGI-3, however, we can see the thought process behind the score, not just the outcome.
However there is still a gap between what humans can learn and what AI can learn. ARC Prize Foundation exists to understand this gap. The ARC-AGI benchmarks are our tools for measuring it.
ARC-AGI-3 is the first fully-interactive benchmark in the ARC-AGI series, featuring hundreds of handcrafted games with thousands of levels.
One Month of Learnings Building Interactive Reasoning Benchmarks
Showing 8 of 8 stories
You're all caught up.
Get the top AI news and tools straight to your inbox every week.
Weekly · AI news
Join 15,000+ AI enthusiasts
