KernelGenBench-MS (Claude Code): leaderboard
KernelGenBench asks models and coding agents to write Triton kernels that drop into PyTorch in place of an existing operator. A kernel counts only when it passes numerical checks across a combinatorial suite of shapes, data types and layouts and clears anti-hack interception that blocks calls back to the reference implementation. This board covers all 210 multi-source operators on an NVIDIA A100, generated by the Claude Code agent with execution feedback.
Metric: Clean pass rate (%; share of the 210 KernelGenBench-MS operators (110 PyTorch ATen, 50 vLLM, 50 cuBLAS) whose generated Triton kernel passes numerical validation on every test case of its combinatorial shape, dtype and layout suite and every anti-hack check, as a drop-in replacement; the Claude Code agent with execution feedback; NVIDIA A100; 30-minute budget per operator). Source: arxiv.org. Saturation forecast: Around February 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 87 | #102 |
| 2 | GLM-5 | 67 | #260 |
| 3 | Qwen 3.5 27B | 62 | #359 |
| 4 | MiniMax-M2.5 | 49 | #582 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=kernelgenbench-ms-claude-code · How It Works · Data refreshed daily, snapshot 2026-11-13.