ARC-AGI-2 (ARC Prize) — leaderboard

Significantly harder successor to ARC-AGI-1, released 2025. Novel grid-based puzzles requiring fluid reasoning and generalization from few examples. Top frontier models score under 5% while humans score ~98%.

Source: arcprize.org. Status: years away from saturation.

Interactive version: theaggregate.ai/benchmark?slug=arc-agi-2-arc-prize · How the rankings work · Data refreshed daily, snapshot 2026-07-22.