Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on HMMT25 (mean@8)
Loading...
13.33
Mean Score (8 samples)
ACPO
4.8748
7.0699
9.265
11.4601
Jul 3, 2026
Mean Score (8 samples)
Updated 18d ago
Evaluation Results
Method
Method
Links
Mean Score (8 samples)
ACPO
Model=DeepSeek-R1-Dist...
2026.07
13.33
GTPO
Model=DeepSeek-R1-Dist...
2026.07
13.19
SAPO
Model=DeepSeek-R1-Dist...
2026.07
12.46
ACPO
Model=Qwen3-8B-Base
2026.07
11.04
80/20
Model=DeepSeek-R1-Dist...
2026.07
10.83
SAPO
Model=Qwen3-8B-Base
2026.07
10
ACPO
Model=Qwen2.5-Math-7B
2026.07
9.86
DAPO
Model=DeepSeek-R1-Dist...
2026.07
9.61
GTPO
Model=Qwen3-8B-Base
2026.07
8.83
GRPO
Model=DeepSeek-R1-Dist...
2026.07
8.75
GTPO
Model=Qwen2.5-Math-7B
2026.07
8.7
SAPO
Model=Qwen2.5-Math-7B
2026.07
8.64
80/20
Model=Qwen3-8B-Base
2026.07
7.91
80/20
Model=Qwen2.5-Math-7B
2026.07
7.7
DAPO
Model=Qwen3-8B-Base
2026.07
7.12
DAPO
Model=Qwen2.5-Math-7B
2026.07
6.67
GRPO
Model=Qwen3-8B-Base
2026.07
6.67
GRPO
Model=Qwen2.5-Math-7B
2026.07
5.2
Feedback
Search any
task
Search any
task