# FastKernels: Benchmarking GPU Kernel Generation in Production

arXiv: https://arxiv.org/abs/2605.23215 · HTML: https://mlsanity.com/paper/2605.23215/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 16 of its 30 references that are in the map.
- First posted on arXiv 22 May 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 16 of 16)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [SGLang: Efficient Execution of Structured Language Model Programs](https://mlsanity.com/paper/2312.07104/) (2023)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [Kevin: Multi-Turn RL for Generating CUDA Kernels](https://mlsanity.com/paper/2507.11948/) (2025)
- [TritonBench: Benchmarking Large Language Model Capabilities for Generating Triton Operators](https://mlsanity.com/paper/2502.14752/) (2025)
- [CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning](https://mlsanity.com/paper/2507.14111/) (2025)
- [Astra: A Multi-Agent System for GPU Kernel Performance Optimization](https://mlsanity.com/paper/2509.07506/) (2025)
- [CudaForge: An Agent Framework with Hardware Feedback for CUDA Kernel Optimization](https://mlsanity.com/paper/2511.01884/) (2025)
- [AutoTriton: Automatic Triton Programming with Reinforcement Learning in LLMs](https://mlsanity.com/paper/2507.05687/) (2025)
- [MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation](https://mlsanity.com/paper/2507.17773/) (2025)
- [FlashInfer-Bench: Building the Virtuous Cycle for AI-driven LLM Systems](https://mlsanity.com/paper/2601.00227/) (2026)
- [Dr. Kernel: Reinforcement Learning Done Right for Triton Kernel Generations](https://mlsanity.com/paper/2602.05885/) (2026)
- [QiMeng-Kernel: Macro-Thinking Micro-Coding Paradigm for LLM-Based High-Performance GPU Kernel Generation](https://mlsanity.com/paper/2511.20100/) (2025)
- [SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits](https://mlsanity.com/paper/2603.19173/) (2026)
- [NPUEval: Optimizing NPU Kernels with LLMs and Open Source Compilers](https://mlsanity.com/paper/2507.14403/) (2025)
- [KernelCraft: Benchmarking for Agentic Close-to-Metal Kernel Generation on Emerging Hardware](https://mlsanity.com/paper/2603.08721/) (2026)

## Built on by (most cited first, 3 of 3)

- [Learning When to Optimize: Verified Optimization Skills from Expert GPU-Kernel Lineages](https://mlsanity.com/paper/2605.28213/) (2026)
- [PTXBench: Benchmarking and Adapting LLMs for GPU Kernel Optimization with Architecture-specific PTX](https://mlsanity.com/paper/2608.17379/) (2026)
- [D2K-Bench: Can LLM Agents Turn Expert Designs into Efficient GPU Kernels?](https://mlsanity.com/paper/2610.03226/) (2026)

## Abstract (arXiv)

LLM-based agents for GPU kernel generation are advancing rapidly, but the benchmarks they optimize against evaluate kernels in isolation, with synthetic inputs and weak baselines, rewarding sandbox speedups that break or vanish in real inference systems. We introduce FastKernels, a benchmark of 384 tasks drawn from 47 representative architectures across 8 categories, whose kernels suffice to reimplement 94.6% (472/499) of HuggingFace Transformers architectures with outputs matching the native implementations. Each task mirrors the interface of the corresponding production module and is scored against the kernels production frameworks ship, and tasks form a compositional hierarchy, from primitives to full models, in which higher-level modules import lower-level ones. Candidates are scored at the kernel level and end to end inside the models they come from, on the production execution path, and MacroEval aggregates calibrated correctness, coverage, and speedup into a leaderboard. Seeding it with five representative agents (6,900 agent-hours), we find that kernel-level speedups of 1.6-6.6× shrink to at most 1.25× end to end, only 20% of winning kernel sets run correctly as-is, and kernel-level scores mis-rank agents: Claude Code matches or beats KDA at every level in isolation, yet KDA scores 3× higher end to end. Code is available at https://github.com/Snowflake-AI-Research/fastkernels.