KernelGenBench-MC (Pass@1) - Iluvatar CoreX: leaderboard
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; direct zero-shot generation, one sample at temperature 0, no execution feedback; an Iluvatar CoreX accelerator; 30-minute budget per operator). Source: arxiv.org. Saturation forecast: Around January 2027. 3 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 4.6 | 38 |
| 2 | Qwen 3.5 27B | 10 |
| 3 | MiniMax-M2.5 | 4 |
Interactive version: theaggregate.ai/benchmark?slug=kernelgenbench-mc-pass-1-iluvatar-corex · How It Works · Data refreshed daily, snapshot 2026-09-29.