Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Software Engineering Efficiency on SWE-bench Lite (Speedup, Mean acceptance length (τ))
Loading...
4.38
Speedup
DFlash+DDTree
2.0088
2.6244
3.24
3.8556
Apr 14, 2026
Speedup
Mean Acceptance Length (τ)
Updated 3d ago
Evaluation Results
Method
Method
Links
Speedup
Mean Acceptance Length (τ)
DFlash+DDTree
Model=Qwen3-Coder-30B-...
2026.04
4.38
5.71
DFlash+DDTree
Model=Qwen3-4B, Temper...
2026.04
4.25
5.99
DFlash+DDTree
Model=Qwen3-8B, Temper...
2026.04
4.23
5.91
DFlash+DDTree
Model=Qwen3-Coder-30B-...
2026.04
3.8
4.96
DFlash+DDTree
Model=Qwen3-4B, Temper...
2026.04
3.71
5.2
DFlash+DDTree
Model=Qwen3-8B, Temper...
2026.04
3.47
4.86
DFlash
Model=Qwen3-Coder-30B-...
2026.04
2.77
3.61
DFlash
Model=Qwen3-4B, Temper...
2026.04
2.7
3.66
DFlash
Model=Qwen3-8B, Temper...
2026.04
2.65
3.6
DFlash
Model=Qwen3-Coder-30B-...
2026.04
2.42
3.16
DFlash
Model=Qwen3-4B, Temper...
2026.04
2.29
3.07
DFlash
Model=Qwen3-8B, Temper...
2026.04
2.1
2.82
Feedback
Search any
task
Search any
task