AccelOpt：面向 AI 加速器内核优化的自我改进型 LLM 智能体系统
阅读原文· arxiv.org研究团队推出 AccelOpt，一种能自主优化 AI 加速器内核的自我改进型 LLM 智能体系统。该系统通过迭代生成和优化记忆库积累经验，无需硬件专家知识即可提升内核性能。在针对 AWS Trainium 构建的 NKIBench 基准测试中，AccelOpt 将 Trainium 1 和 Trainium 2 的平均峰值吞吐量分别从 49% 和 45% 提升至 61% 和 59%。该系统采用开源模型即可达到与 Claude Sonnet 4 相当的优化效果，成本却降低 26 倍。
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 26times cheaper. The code is open-sourced at https://github.com/zhang677/AccelOpt.