Math Benchmarks
Benchmarks
TaskDatasetSOTA resultTrendResults
Math Benchmarks Aggregate
81.9Accuracy (Avg)
62
Math Benchmarks GSM8K, Minerva, MATH, MathQA
59.24GSM8K Score
53
Math Benchmarks Average
76.1Accuracy (ACC)
47
Math Benchmarks Aggregate
71.8Pass@1
44
Math Benchmarks AIME 2024 AIME 2025 OlympiadBench
19.5AIME 2024 Score
19
Math Benchmarks (MATH, Minerva, Olympiad, AIME24, AIME25, AMC23)
70.65MATH Score
18
Mean of six math benchmarks
43.8Pass@1
12
Math Benchmarks Overall (test)
87Pass@1
12
Math Benchmarks LIMO curation (test)
72.6Accuracy
10
Math Benchmarks MATH, GSM8K, AMC23, AIME24, Minerva, Gaokao, Olympiad (test)
75.1MATH Score
10
Math Benchmarks Task-average
42.3Pass@1
9
Math Benchmarks (GSM8K, MATH, AMC23, AIME24) (test)
96Accuracy (GSM8K)
8
Math Benchmarks Math500, OlympiadBench, Minerva, AIME, AMC
85.6Math500 Accuracy
7
Math Benchmarks evaluated on Llama 3-70B
78.2GSM8K Accuracy
5
Math Benchmarks MATH, MATH500, ThmQA
67.6MATH multi@5 Accuracy
4
Math Benchmarks (test)
28.9GSM8K Accuracy
3