Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
CUDA Code Validation on KernelBench Level 1
Loading...
8
Success Rate (p=1.0)
qwen2.5-7b
-0.32
1.84
4
6.16
Dec 4, 2025
Success Rate (p=1.0)
Success Rate (p=1.5)
Success Rate (p=2.0)
Geometric Mean Speedup
Updated 4d ago
Evaluation Results
Method
Method
Links
Success Rate (p=1.0)
Success Rate (p=1.5)
Success Rate (p=2.0)
Geometric Mean Speedup
qwen2.5-7b
Model=qwen2.5-7b, Simu...
2025.12
8
2
1
99.2
qwen2.5-72b
Model=qwen2.5-72b
2025.12
6
3
2
20.1
qwen2.5-7b
Model=qwen2.5-7b, Simu...
2025.12
5
2
1
100.1
llama3.1-405b
Model=llama3.1-405b
2025.12
2
2
2
11.2
qwen2.5-7b
Model=qwen2.5-7b
2025.12
0
0
0
22.7
Feedback
Search any
task
Search any
task