KernelGenBench-MC (Claude Code) - Platform A: 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 the 110 PyTorch ATen operators on an anonymized accelerator platform, generated by the Claude Code agent with execution feedback.
Metric: Clean pass rate (%; share of the 110 PyTorch ATen operators 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; de-identified Platform A; 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 | 89 | #102 |
| 2 | Qwen 3.5 27B | 78 | #359 |
| 3 | MiniMax-M2.5 | 69 | #582 |
| 4 | GLM-5 | 65 | #260 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=kernelgenbench-mc-claude-code-platform-a · How It Works · Data refreshed daily, snapshot 2026-11-13.