Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Math Reasoning on AIME 2024 (Pass@1, Pass@16)
Loading...
41.66
Pass@1
EAPO
31.2704
33.9677
36.665
39.3623
Jun 29, 2026
Pass@1
Pass@16
Updated 25d ago
Evaluation Results
Method
Method
Links
Pass@1
Pass@16
EAPO
Model=Qwen-3-8B-Base
2026.06
41.66
74.37
EAPO
Model=Qwen-2.5-Math-7B
2026.06
40.14
60.44
EAPO w/o sIS
Model=Qwen-2.5-Math-7B
2026.06
39.68
63.29
MOPD
Model=Qwen-2.5-Math-7B
2026.06
38.85
60.28
OPD
Model=Qwen-3-8B-Base
2026.06
38.23
67.48
OPD
Model=Qwen-2.5-Math-7B
2026.06
38.12
55.06
MOPD
Model=Qwen-3-8B-Base
2026.06
37.8
74.68
Trajectory Replay
Model=Qwen-2.5-Math-7B
2026.06
37.25
58.6
DAPO
Model=Qwen-2.5-Math-7B
2026.06
37.19
58
DAPO
Model=Qwen-3-8B-Base
2026.06
36.67
71.41
GRPO
Model=Qwen-2.5-Math-7B
2026.06
32.08
50.24
GRPO
Model=Qwen-3-8B-Base
2026.06
31.67
61.98
Feedback
Search any
task
Search any
task