Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Generation on MBPP, CodeAlpacaPy, HumanEval, and LiveCodeBench (Composite)
Loading...
4.04
Speedup
DEER
1.1176
1.8763
2.635
3.3937
Dec 17, 2025
Speedup
Avg Acceptance Length (τ)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Speedup
Avg Acceptance Length (τ)
DEER
Base Model=Qwen3-30B-A...
2025.12
4.04
5.03
DEER
Base Model=Qwen3-14B,...
2025.12
2.98
4.82
DEER
Base Model=Qwen3-8B, T...
2025.12
2.83
4.61
DEER
Base Model=Qwen3-4B, T...
2025.12
2.77
4.61
DEER
Base Model=Qwen2-7B, T...
2025.12
2.7
4.11
EAGLE3
Base Model=Qwen3-8B, T...
2025.12
2.48
3.45
EAGLE3
Base Model=Qwen2-7B, T...
2025.12
2.43
3.22
EAGLE3
Base Model=Qwen3-4B, T...
2025.12
2.4
3.31
EAGLE3
Base Model=Qwen3-14B,...
2025.12
2.39
3.54
Hydra
Base Model=Qwen3-8B, T...
2025.12
2.25
2.66
Hydra
Base Model=Qwen3-4B, T...
2025.12
2.22
2.58
EAGLE3
Base Model=Qwen3-30B-A...
2025.12
2.21
3.05
MEDUSA
Base Model=Qwen3-8B, T...
2025.12
1.32
1.97
MEDUSA
Base Model=Qwen3-4B, T...
2025.12
1.23
1.94
Feedback
Search any
task
Search any
task
