Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
22.2Final Accuracy
2
Mar 4, 2026
16.67Final Accuracy
2
Mar 4, 2026
9.8Final Accuracy
2
Mar 4, 2026
37.83Throughput
2
Feb 26, 2026
98.56Pass@1 Accuracy
2
Feb 26, 2026
83.8Pass@50
2
Feb 26, 2026
52.26Score
2
Feb 26, 2026
93.37Score
2
Feb 26, 2026
3.55Inference Cost (USD)
2
Feb 26, 2026
5.83Inference Cost (USD)
2
Feb 26, 2026
0.056MAE
2
Feb 26, 2026
2.12Speedup
2
Feb 26, 2026
2.23Speedup
2
Feb 26, 2026
1.44GSM8k Accuracy
2
Feb 26, 2026
95.6Accuracy
2
Feb 26, 2026
41Score
2
Feb 26, 2026
78.8Accuracy
2
Feb 26, 2026
77Accuracy
2
Feb 26, 2026
0.211Accuracy
2
Feb 26, 2026
20.8Accuracy
2
Feb 26, 2026
50Avg@10
2
Feb 26, 2026
56.7Accuracy
2
Feb 26, 2026
39.8Accuracy
2
Feb 26, 2026
51.42Pass@4
2
Feb 26, 2026
34.28Pass@4
2
Feb 26, 2026