Links

Schema solves ARC-AGI-3 Public

Benchmark
https://schema-harness.github.io

If the score holds on the private dataset, Schema demonstrates models have latent intelligence unlocked through better harness engineering.

Schema scores 98.98 and 95.35 percent on the ARC-AGI-3 public evaluation setSchema scores 98.98 and 95.35 percent on the ARC-AGI-3 public evaluation set

Published on July 17, 2026