Abstraction and Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
397Accuracy
13
Jun 26, 2026
100Pass@2
13
Apr 6, 2026
0.41Total Cost
10
Feb 26, 2026
58.2ARC-1 Score
9
Mar 30, 2026
68.2RSR (%)
5
May 15, 2026
7.8Pass@2
4
Jun 17, 2026
47.5Pass@2
4
Jun 17, 2026