Kernel Generation on KernelBench AVG Level-1 2 3
99.33Correlation CoefficientClaude-4.6-opus
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Claude-4.6-opusFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 99.33 | 77 | 3.33 | |
| Kernel-Smith-235B-RLFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 96.33 | 70 | 3.7 | |
| DeepSeek-v3.2-SpecialeFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 94.67 | 61 | 3.44 | |
| Gemini-3.0-proFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 94.33 | 74 | 2.83 | |
| Qwen3-235B-2507-thinkFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 90.67 | 62 | 2.2 | |
| Qwen3.5-397B-thinkFramework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 84.33 | 55 | 2.16 | |
| Kimi-K2.5Framework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 84.33 | 55 | 2.16 | |
| Minimax-M2.5Framework=Kernel-Smith evolution-agent, Evolution Rounds=402026.03 | 72.33 | 52 | 1.27 |