Longterm Wiki

ARC-AGI

Reasoning

Abstraction and Reasoning Corpus — a benchmark of visual pattern recognition tasks designed to test fluid intelligence and novel reasoning.

Models Tested
2
Best Score
87.5%
Median Score
85.05%
Scoring: accuracy
Introduced: 2019-11
Maintainer: Francois Chollet

Leaderboard2 models

#ModelDeveloperScore
🥇o3OpenAI87.5%
🥈o4-miniOpenAI82.6%