Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematics on AIME 25 (Req/s, Speedup, Acc.)
Loading...
39.43
Req/s
SQUEEZE EVOLVE (p=0)
-0.1628
10.1161
20.395
30.6739
Apr 9, 2026
Req/s
Speedup
Accuracy
Updated 9d ago
Evaluation Results
Method
Method
Links
Req/s
Speedup
Accuracy
SQUEEZE EVOLVE (p=0)
Model 1=GPT-OSS-20B, M...
2026.04
39.43
2.31
90.8
SQUEEZE EVOLVE (p=10)
Model 1=GPT-OSS-20B, M...
2026.04
24.59
1.44
90.5
RSA
Model 1=GPT-OSS-20B, M...
2026.04
17.09
1
90.1
SQUEEZE EVOLVE (p=0)
Model 1=Qwen3-30B-A3B-...
2026.04
13.47
9.9
81
SQUEEZE EVOLVE (p=10)
Model 1=Qwen3-30B-A3B-...
2026.04
7.41
5.44
80.1
RSA
Model 1=Qwen3-30B-A3B-...
2026.04
1.36
1
82
Feedback
Search any
task
Search any
task