Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Generation on HumanEval (MAT, Speed)
Loading...
9.48
MAT
TALON
2.512
4.321
6.13
7.939
Jan 12, 2026
MAT
Speed
Updated 1mo ago
Evaluation Results
Method
Method
Links
MAT
Speed
TALON
Model=Vicuna-13B, Temp...
2026.01
9.48
5.16
EAGLE-3
Model=Vicuna-13B, Temp...
2026.01
8.31
4.77
OPT-Tree
Model=Vicuna-13B, Temp...
2026.01
8.21
4.35
TALON
Model=Llama3-8B, Tempe...
2026.01
7.28
4.2
EAGLE-3
Model=Llama3-8B, Tempe...
2026.01
7.08
4.05
EAGLE-3
Model=DSL-8B, Temperat...
2026.01
6.7
3.85
TALON
Model=DSL-8B, Temperat...
2026.01
6.28
3.96
EAGLE-3
Model=Qwen3-8B, Temper...
2026.01
3.89
2.35
HYDRA
Model=Vicuna-13B, Temp...
2026.01
3.87
2.67
TALON
Model=Qwen3-8B, Temper...
2026.01
3.78
2.69
EAGLE-3
Model=Qwen3-32B, Tempe...
2026.01
2.98
1.96
TALON
Model=Qwen3-32B, Tempe...
2026.01
2.96
2.15
SD
Model=Vicuna-13B, Temp...
2026.01
2.86
1.47
MEDUSA
Model=Vicuna-13B, Temp...
2026.01
2.78
2.18
Feedback
Search any
task
Search any
task