MATH
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
MATH 500
96.6MATH 500 Score
25
MATH 500
92.7Accuracy
25
Math
1Average Similarity
25
MATH 500
94Overall Score
25
MATH-500
73.6Accuracy
25
MATH-500
82.65Pass@1
25
MATH 500
6.17Speedup Ratio
24
MATH 500
97.4Accuracy
24
MATH-500
97.2Maj@16
24
MATH 500
7.64Speedup
24
MATH 500
91.3Mean@16 Accuracy
24
MATH 500
95.66Accuracy
24
MATH 500
79.3Pass@k
24
MATH500
85Accuracy
24
MATH 4-shot
53.8NFE (4-shot)
24
MATH 500
95.2Accuracy
24
Math GSM8K AIME24
46.9Accuracy (GSM8K)
24
MATH-P
48.4Pass@1
24
MATH (test)
48.67Exact Match (EM)
24
MATH500
78Pass@k Rate
24
MATH500
81Accuracy
24
MATH-500
93Accuracy
24
MATH500
88.05Avg@16 Score
24
MATH500
1.2Inference Latency (s)
24
MATH500
12Memory Usage Ratio
24