Schema Harness Achieves ~99% on Arc‑AGI‑3 Public
A new framework called Schema has significantly improved the performance of AI models on the ARC-AGI-3 benchmark. By optimizing the process of how models interact with game environments, it achieved a 99% success rate on public sets.
Why it matters
This demonstrates that architectural process improvements can yield massive performance gains in AI reasoning without needing to retrain underlying model weights.
ARC‑AGI‑3 gives an agent a game environment, without an explanation of what it is seeing. At each step, the agent receives a 64×64 grid of 16 color indices and a set of legal actions. The environment supplies no object list, rule sheet, stated goal, or shaped reward. There is only one way to make progress: the physicist’s way. The agent must act while its model of the game is still provisional, forming hypotheses about what the grid represents, how actions change it, and what counts as success, then revising both its model and its plan as new observations arrive.
Technical reporting on AI benchmarks with no political or social agenda.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in