Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
93.49Pass@4
10
May 7, 2026
98.44Pass@4 Success Rate
10
May 7, 2026
49.38Accuracy
10
May 7, 2026
72.03Accuracy
10
May 7, 2026
258.35Decoding Speed (tokens/s)
10
May 4, 2026
4.3Reward-weighted Pass@1
10
Apr 24, 2026
3.45Reward-weighted Pass@1
10
Apr 24, 2026
32.01Overall Score
10
Apr 21, 2026
70.4Accuracy
10
Apr 21, 2026
62.1CoT-E
10
Apr 21, 2026
76.67Accuracy (MATH)
10
Apr 20, 2026
72.6Average Accuracy
10
Apr 16, 2026
93.2Accuracy (avg@32)
10
Apr 16, 2026
48.6Average Score
10
Apr 14, 2026
45.5Avg@4 Score
10
Apr 14, 2026
63.8Avg@32
10
Apr 14, 2026
17.2Average Score (Avg@32)
10
Apr 14, 2026
49.7Average@32
10
Apr 14, 2026
89.8Average Score (Avg@32)
10
Apr 14, 2026
30.4Avg@32
10
Apr 14, 2026
42.7Avg@32
10
Apr 14, 2026
69.5Pass@1
10
Apr 14, 2026
35.2Pass@1
10
Apr 14, 2026
42.9Pass@1
10
Apr 14, 2026
60.8Pass@1
10
Apr 14, 2026