Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Math Reasoning on AIME 2025 (test) (pass@1)
Loading...
50
Pass@1 Rate
Merged (α=0.2)
22.232
29.441
36.65
43.859
Apr 12, 2026
Pass@1 Rate
Updated 3mo ago
Evaluation Results
Method
Method
Links
Pass@1 Rate
Merged (α=0.2)
Model variant=Merged (...
2026.04
50
Qwen3-4B-Instruct-2507
Model variant=Base Model
2026.04
46.7
SFT No-Test
Model variant=SFT No-Test
2026.04
23.3
Feedback
Search any
task
Search any
task