Schema Harness Claims ~99% Score on ARC-AGI-3 Public Benchmark
A system called Schema Harness has reportedly achieved approximately 99% accuracy on the ARC-AGI-3 Public benchmark, a challenging test designed to measure general reasoning abilities in AI. The result was shared on Hacker News, where it attracted modest early attention with 18 points and 2 comments. ARC-AGI benchmarks are widely regarded as difficult evaluations of abstract reasoning, making a near-perfect score a notable claim. The project has a dedicated website where further details about the approach are presumably documented. The submission provides limited context, and the claim has not yet been independently verified or widely discussed.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in