Zero-shot Evaluation
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
45ARC-C Accuracy
6
Apr 21, 2026
47.7ARC Challenge Accuracy
6
Apr 14, 2026
61.73LAMBDA Score
6
Mar 30, 2026
60.65ARC-E Accuracy
6
Mar 11, 2026
57.90-shot Score
6
Feb 26, 2026
69.82Average Score
5
Jun 4, 2026
70.97Accuracy
5
May 25, 2026
31.63Arc-e Accuracy
5
Feb 26, 2026
-0.93Avg ∆Acc (%)
4
May 21, 2026
29.23Hellaswag Accuracy (Zero-shot)
4
Apr 15, 2026
70.8Accuracy (Zero-shot Non-reasoning)
4
Feb 26, 2026
55.5Average Score (7-task Suite)
3
Jun 17, 2026
44.4Overall Accuracy
3
Feb 26, 2026