Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911618093285376 |
|---|---|
| author | Du, He Ge, Qiming Hu, Jiakai Yang, Aijun Cai, Zheng Huang, Zixian Yuan, Sheng Cheng, Qinxiu Xie, Xinchen Chen, Yicheng Li, Yining Xie, Jiaxing Dong, Huanan Wu, Yaguang Huang, Xiangjun Yang, Jian Wang, Hui Zhou, Bowen Li, Bowen Guo, Qipeng Chen, Kai |
| author_facet | Du, He Ge, Qiming Hu, Jiakai Yang, Aijun Cai, Zheng Huang, Zixian Yuan, Sheng Cheng, Qinxiu Xie, Xinchen Chen, Yicheng Li, Yining Xie, Jiaxing Dong, Huanan Wu, Yaguang Huang, Xiangjun Yang, Jian Wang, Hui Zhou, Bowen Li, Bowen Guo, Qipeng Chen, Kai |
| contents | We present Kernel-Smith, a framework for high-performance GPU kernel and operator generation that combines a stable evaluation-driven evolutionary agent with an evolution-oriented post-training recipe. On the agent side, Kernel-Smith maintains a population of executable candidates and iteratively improves them using an archive of top-performing and diverse programs together with structured execution feedback on compilation, correctness, and speedup. To make this search reliable, we build backend-specific evaluation services for Triton on NVIDIA GPUs and Maca on MetaX GPUs. On the training side, we convert long-horizon evolution trajectories into step-centric supervision and reinforcement learning signals by retaining correctness-preserving, high-gain revisions, so that the model is optimized as a strong local improver inside the evolutionary loop rather than as a one-shot generator. Under a unified evolutionary protocol, Kernel-Smith-235B-RL achieves state-of-the-art overall performance on KernelBench with Nvidia Triton backend, attaining the best average speedup ratio and outperforming frontier proprietary models including Gemini-3.0-pro and Claude-4.6-opus. We further validate the framework on the MetaX MACA backend, where our Kernel-Smith-MACA-30B surpasses large-scale counterparts such as DeepSeek-V3.2-think and Qwen3-235B-2507-think, highlighting potential for seamless adaptation across heterogeneous platforms. Beyond benchmark results, the same workflow produces upstream contributions to production systems including SGLang and LMDeploy, demonstrating that LLM-driven kernel optimization can transfer from controlled evaluation to practical deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_28342 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization Du, He Ge, Qiming Hu, Jiakai Yang, Aijun Cai, Zheng Huang, Zixian Yuan, Sheng Cheng, Qinxiu Xie, Xinchen Chen, Yicheng Li, Yining Xie, Jiaxing Dong, Huanan Wu, Yaguang Huang, Xiangjun Yang, Jian Wang, Hui Zhou, Bowen Li, Bowen Guo, Qipeng Chen, Kai Computation and Language Machine Learning We present Kernel-Smith, a framework for high-performance GPU kernel and operator generation that combines a stable evaluation-driven evolutionary agent with an evolution-oriented post-training recipe. On the agent side, Kernel-Smith maintains a population of executable candidates and iteratively improves them using an archive of top-performing and diverse programs together with structured execution feedback on compilation, correctness, and speedup. To make this search reliable, we build backend-specific evaluation services for Triton on NVIDIA GPUs and Maca on MetaX GPUs. On the training side, we convert long-horizon evolution trajectories into step-centric supervision and reinforcement learning signals by retaining correctness-preserving, high-gain revisions, so that the model is optimized as a strong local improver inside the evolutionary loop rather than as a one-shot generator. Under a unified evolutionary protocol, Kernel-Smith-235B-RL achieves state-of-the-art overall performance on KernelBench with Nvidia Triton backend, attaining the best average speedup ratio and outperforming frontier proprietary models including Gemini-3.0-pro and Claude-4.6-opus. We further validate the framework on the MetaX MACA backend, where our Kernel-Smith-MACA-30B surpasses large-scale counterparts such as DeepSeek-V3.2-think and Qwen3-235B-2507-think, highlighting potential for seamless adaptation across heterogeneous platforms. Beyond benchmark results, the same workflow produces upstream contributions to production systems including SGLang and LMDeploy, demonstrating that LLM-driven kernel optimization can transfer from controlled evaluation to practical deployment. |
| title | Kernel-Smith: A Unified Recipe for Evolutionary Kernel Optimization |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2603.28342 |