Skip to content
Longterm Wiki has not been actively maintained since May 2026. Some data is still updated occasionally, but please don't rely on this site being current or accurate.
Longterm Wiki

ARC-AGI-2

Reasoning
Wiki pageWebsiteData
Second iteration of the ARC benchmark with harder tasks, designed to remain challenging as AI capabilities improve.
Models Tested
2
Best Score
77.1
Median Score
72.95
Scoring: accuracy
Introduced: 2025-01
Maintainer: Francois Chollet / ARC Prize Foundation

Leaderboard (2 models)

#ModelDeveloperScore
🥇GeminiGoogle DeepMind
77.1
🥈Claude Opus 4.6Anthropic
68.8%