Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
51.1Pass@1
12
May 29, 2026
23.3Pass@1
12
May 29, 2026
42.5Pass@1
12
May 29, 2026
56.5Pass@1
12
May 29, 2026
2.2Latency (s)
12
May 29, 2026
2.3Latency (s)
12
May 29, 2026
93Pass@k
12
May 28, 2026
1.4Average@10
12
May 28, 2026
3.6Average Score (a@10)
12
May 28, 2026
21.9Average@10
12
May 28, 2026
48.5Average@10 Score
12
May 28, 2026
44Average@10
12
May 28, 2026
68.2Average Score @10
12
May 28, 2026
173Avg Training Steps to Target Acc
12
May 27, 2026
180Training Steps to Target Accuracy
12
May 27, 2026
220Steps to Target Accuracy
12
May 27, 2026
160Training Steps to Target Accuracy
12
May 27, 2026
160Training Steps to Target Accuracy
12
May 27, 2026
100Training Steps
12
May 27, 2026
20Training Steps
12
May 27, 2026
91.5Accuracy (Avg@8)
12
May 27, 2026
68.7Avg@8 Success Rate
12
May 27, 2026
84.2Peak Accuracy
12
May 27, 2026
0.57Peak Accuracy
12
May 27, 2026
39.3Peak Accuracy
12
May 27, 2026