Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
71.8Mean@10
4
Feb 26, 2026
36.5Score
4
Feb 26, 2026
82.94GSM8K
4
Feb 26, 2026
65Accuracy
4
Feb 26, 2026
29.8Accuracy
4
Feb 26, 2026
22.8Accuracy
4
Feb 26, 2026
74.4Accuracy
4
Feb 26, 2026
89.47Throughput
4
Feb 26, 2026
41.6Accuracy
4
Feb 27, 2026
82.03ACC
4
Feb 26, 2026
43,638Tokens per Correct
4
Feb 26, 2026
100pass@1
4
Feb 26, 2026
54.5Pass@1 (1K Budget)
4
Feb 26, 2026
30Pass@1
4
Feb 26, 2026
4.51Speedup
4
Feb 26, 2026
4.64Speedup
4
Feb 26, 2026
76Score
4
Mar 27, 2026
33.33Pass@1 Accuracy
4
Feb 26, 2026
66.67Pass@16
4
Feb 26, 2026
85.9Accuracy
4
Feb 26, 2026
43Accuracy
4
Feb 26, 2026
80.97Accuracy
4
Feb 26, 2026
78.43Accuracy
4
Feb 26, 2026
50Accuracy
4
Feb 26, 2026
78.4HS
4
Feb 26, 2026