Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks
By: Jianghui Wang , Vinay Joshi , Saptarshi Majumder and more
Potential Business Impact:
AI writes faster computer code for graphics cards.
The demand for AI-generated GPU kernels is rapidly growing, influenced by the need for scalable, hardware-optimized solutions in both industry and academia. As deep learning workloads grow in complexity and diversity, it is imperative to automate low-level kernel development to meet performance and productivity demands. Major cloud providers, semiconductor companies, and research institutions are now investing heavily in AI-driven code generation for GPUs, aiming to reduce manual optimization efforts while achieving near-expert performance on hardware like AMD MI300X. The Triton language, a Python-based DSL for GPU programming, has emerged as a popular target for such AI-generated kernels due to its balance of performance and ease-of-coding. In this work, we present an evaluation suite for Triton-based GPU kernels and GEAK (Generating Efficient AI-centric GPU Kernels)-a framework that leverages cutting-edge LLMs to generate performant Triton code specifically for AMD GPUs, including the AMD MI300X and MI250. GEAK leverages inference-time compute scaling to produce Triton-based GPU kernels using a reasoning loop adapted from Reflexion-style feedback mechanisms. On two evaluation benchmarks, GEAK significantly outperformed the baselines of directly prompting frontier LLMs as well as Reflexion-based generation pipelines by achieving correctness up to $63$% and execution speed up of up to $2.59$X. These results highlight the promise of GEAK-like agentic code generation for accelerating the adoption of diverse hardware platforms and democratizing access to expert-level kernel performance.
Similar Papers
AKG kernel Agent: A Multi-Agent Framework for Cross-Platform Kernel Synthesis
Artificial Intelligence
AI writes code to make computer brains faster.
TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization
Software Engineering
Makes computer graphics run much faster automatically.
TritonForge: Profiling-Guided Framework for Automated Triton Kernel Optimization
Software Engineering
Makes computer programs run much faster automatically.