Benchmark · reasoning

ARC-AGI 3

1 条结果 1 个模型

ARC Prize's third generation (2026): the agent must learn an unseen interactive game environment from scratch. Scores run far lower than — and are NOT comparable to — the static grid-puzzle ARC-AGI 1/2.

0 25 50 75 100 2026-07-17 Schema · 99 · 2026-07-17
Schema
时间线
日期 模型 得分 来源
2026-07-17 Schema 99.0% Moonshot发布Kimi K3;Schema harness在ARC-AGI 3上达到99%