AIME
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
AIME 2024
80Accuracy
15
AIME 2025
72.5Average@12 Accuracy
15
AIME 2024
80.56Average@12 Accuracy (AIME 2024)
15
AIME25
19.38Avg@k
15
AIME 24
15.21Avg@k
15
AIME 25
61.46Accuracy (AIME 25)
15
AIME 2025
60.42Average Score (32 samples)
15
AIME 2025
16.51Mean Score (32 Samples)
15
AIME 2024
19.9Mean Score (k=32)
15
AIME 25
26.7Pass@1
15
AIME 2025
72.2Average Score
15
AIME 2024
79.2Avg@12 Score
15
AIME Complex Samples 24/25
23.86AVG@8
15
AIME Simple Samples 24/25
99.75AVG@8 Success Rate
15
AIME 2025
44.6Avg@16 Accuracy
15
AIME 2024
65.8Avg Accuracy @16
15
AIME 2025
36Avg@10 Accuracy
15
AIME 2024
39.33Avg@10 Accuracy
15
AIME 2025
72.2Pass@1 Accuracy
15
AIME 25
30.4Accuracy
15
AIME 2025
64.79Accuracy
15
AIME 25
24.17AIME 25 Accuracy
15
AIME 24
28.54Accuracy
15
AIME25
15.6Accuracy (avg@32)
15
AIME 2024
31.8Average@16
15