paper-with-me

Papers

AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization

2025-11-19 · Genghan Zhang, Shaowei Zhu, Anjiang Wei, Zhenyu Song, Allen Nie, Zhen Jia, Nandita Vijaykumar, Yida Wang, Kunle Olukotun arxiv

We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerators, eliminating the need for expert-provided hardware-specific optimization knowledge. AccelOpt explores the kernel optimization space through iterative generation, informed by an optimization memory that curates experiences and insights from previously encountered slow-fast kernel pairs. We build NKIBench, a new benchmark suite of AWS Trainium accelerator kernels with varying complexity extracted from real-world LLM workloads to evaluate the effectiveness of AccelOpt. Our evaluation confirms that AccelOpt's capability improves over time, boosting the average percentage of peak throughput from $49\%$ to $61\%$ on Trainium 1 and from $45\%$ to $59\%$ on Trainium 2 for NKIBench kernels. Moreover, AccelOpt is highly cost-effective: using open-source models, it matches the kernel improvements of Claude Sonnet 4 while being $26\times$ cheaper. The code is open-sourced at https://github.com/zhang677/AccelOpt.

📄 PDF Abstract BibTeX arXiv:2511.15915

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MaxKernel: Agentic Kernel Generation for TPUs

2026-09-03 · Shangkun Wang, Nina Cai, Charles Hoong, Julian Walker 외 arxiv

Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-level expertise. Large Language Models (LLM) can be leveraged together with real-time compiler feedba…

KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware

2026-02-10 · Jiayi Nie, Haoran Wu, Yao Lai, Zeyu Cao 외 arxiv

New AI accelerators with novel instruction set architectures (ISAs) often require developers to manually craft low-level kernels, a time-consuming and error-prone process that does not scale across hardware targets. This…

Code Generation

KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta

2025-12-29 · Gang Liao, Hongsen Qin, Ying Wang, Alicia Golden 외 arxiv

Making deep learning recommendation model (DLRM) training and inference fast and efficient is important. However, this presents three key system challenges - model architecture diversity, kernel primitive diversity, and …

KernelGenBench: Can LLMs and Agents Write Efficient Kernels Across Operator Sources and Hardware Platforms?

2026-07-22 · Peiyu Zang, Jian Tao, Jialing Zhang, Yichen Yuan 외 arxiv

Modern AI systems depend on specialized accelerator kernels, whose development is complicated by increasingly diverse operators and hardware. LLMs and agentic systems promise to automate this work, but existing evaluatio…

Towards Agentic AI on Particle Accelerators

2024-09-10 · Antonin Sulc, Thorsten Hellert, Raimund Kammering, Hayden Hoschouer 외

As particle accelerators grow in complexity, traditional control methods face increasing challenges in achieving optimal performance. This paper envisions a paradigm shift: a decentralized multi-agent framework for accel…