Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Mathematical Reasoning on AIME 25 (Accepted length (τ))
Loading...
5.12
Accepted Length (τ)
DSpark
3.6536
4.0343
4.415
4.7957
Jul 6, 2026
Accepted Length (τ)
Updated 18d ago
Evaluation Results
Method
Method
Links
Accepted Length (τ)
DSpark
target_model=Gemma4-12B
2026.07
5.12
DSpark
target_model=Qwen3-8B
2026.07
5.01
DSpark
target_model=Qwen3-14B
2026.07
4.94
DSpark
target_model=Qwen3-4B
2026.07
4.89
Eagle3
target_model=Gemma4-12B
2026.07
4.83
DFlash
target_model=Gemma4-12B
2026.07
4.22
DFlash
target_model=Qwen3-4B
2026.07
4.15
DFlash
target_model=Qwen3-8B
2026.07
4.07
DFlash
target_model=Qwen3-14B
2026.07
3.98
Eagle3
target_model=Qwen3-4B
2026.07
3.92
Eagle3
target_model=Qwen3-8B
2026.07
3.91
Eagle3
target_model=Qwen3-14B
2026.07
3.71
Feedback
Search any
task
Search any
task