KernelBench-Verified - Level 2: leaderboard
Metric: Fast@1 (%; 97 active fused-operator problems, three constant-output problems excluded; share of problems whose fastest correct sample beats the baseline; single-turn KernelBench prompt, best of five samples; correctness gated on four hidden input distributions (original, x3, x0.01, negated); timing against a TF32-enabled eager PyTorch baseline on an H200). Source: arxiv.org. Saturation forecast: Around January 2027. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 64.9 |
| 2 | Gemini 3.1 Pro (Preview) | 62.9 |
| 3 | Claude Opus 4.8 | 58.8 |
| 4 | Gemini 3 Flash (Preview) | 58.8 |
| 5 | Kimi K2.6 | 54.6 |
| 6 | Claude Opus 4.7 | 52.6 |
| 7 | Claude Sonnet 4.6 | 48.5 |
Interactive version: theaggregate.ai/benchmark?slug=kernelbench-verified-level-2 · How It Works · Data refreshed daily, snapshot 2026-09-29.