Skip to content
AI Atlas
BenchmarkActivecategory · reasoningfamily · arc-agi · variant 2

ARC-AGI-2

arcprize.org/arc-agi/2

harder abstraction puzzles released in 2025 (ARC-AGI-2), designed to resist brute-force program search

data quality57

Updated 2 h ago · first seen 12 Sept 2026

Metric
accuracy · %
Current results
0
Models
0
Current leader

No results recorded for this benchmark yet — its sources are being connected. The definition, aliases and variants are kept so links resolve; nothing is fabricated.

Leaderboard 0 models

Select models with +, then Compare.

No result in this group with these filters

Relax the trust / organization filters or pick another comparability group.