Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
66.7Maj@32
8
Jun 9, 2026
60Maj@32
8
Jun 9, 2026
66.7Majority Score @32
8
Jun 9, 2026
65Maj@16
8
Jun 9, 2026
16.7Majority Success Rate @32
8
Jun 9, 2026
14Bound Accuracy
8
Jun 8, 2026
40.7Pass@1
8
Jun 5, 2026
13.76ECE
8
Jun 5, 2026
86.09Accuracy
8
Jun 5, 2026
38.12Accuracy
8
Jun 5, 2026
40Accuracy
8
Jun 5, 2026
84.8Pass@1 Final
8
Jun 5, 2026
82.1Accuracy (Avg@8)
8
Jun 5, 2026
62.7GSM8K Accuracy
8
Jun 4, 2026
4.95Latency (ms/tok)
8
Jun 2, 2026
4.53Latency (ms/tok)
8
Jun 2, 2026
92.94Accuracy (L1)
8
Jun 2, 2026
2.62Speedup Ratio
8
Jun 2, 2026
71.5Expert Count
8
Jun 1, 2026
70Expert Count
8
Jun 1, 2026
85.83Acc (AMC23)
8
May 29, 2026
90Pass@1 Accuracy
8
May 29, 2026
38.36Exact Match Accuracy
8
May 28, 2026
66.6Avg@4 Score
8
May 27, 2026
30.07Average Performance
8
May 26, 2026