Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Math generation on AIME
Loading...
33.07
Pass@10
Qwen2.5-VL-72B-Ins
8.318
14.744
21.17
27.596
Sep 29, 2025
Pass@10
Pass@1
Pass@32
GSS@10
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass@10
Pass@1
Pass@32
GSS@10
Qwen2.5-VL-72B-Ins
2025.09
33.07
-
41.39
-
GRPO-MA
TN=4, AN=4, S/S=132.87
2025.09
14.7
3.4
27.6
5
GRPO
TN=16, AN=1, S/S=225.43
2025.09
12.89
2.93
21.72
15
GRPO
TN=4, AN=1, S/S=111.24
2025.09
11.78
2.07
20.32
5
GRPO
TN=8, AN=1, S/S=140.05
2025.09
11.16
2.2
21.3
13
SFT
2025.09
11.07
-
18.11
-
Qwen2.5-VL-7B-Ins
2025.09
9.97
-
18.39
-
Qwen2.5-VL-3B-Ins
2025.09
9.27
-
16.25
-
Feedback
Search any
task
Search any
task