Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Code Generation on HumanEval (Accuracy, Rank)
Loading...
20
Accuracy
Top-k Sampling
17.92
18.46
19
19.54
May 10, 2026
Accuracy
Rank
Updated 22d ago
Evaluation Results
Method
Method
Links
Accuracy
Rank
Top-k Sampling
Model Family=Gemma
2026.05
20
2.63
Top-k Sampling
Model Family=Granite
2026.05
20
4.75
Beam Search (3)
Model Family=Gemma
2026.05
19
2.33
EDEN
Model Family=Gemma
2026.05
19
2
Beam Search (3)
Model Family=Granite
2026.05
19
2.5
EDEN
Model Family=Granite
2026.05
19
1.37
Greedy
Model Family=Gemma
2026.05
18
4.25
Top-p Sampling
Model Family=Gemma
2026.05
18
4.25
Min-p Sampling
Model Family=Gemma
2026.05
18
4.25
Greedy
Model Family=Granite
2026.05
18
4.25
Top-p Sampling
Model Family=Granite
2026.05
18
4.5
Min-p Sampling
Model Family=Granite
2026.05
18
3.62
Feedback
Search any
task
Search any
task