GPU Kernel Generation

GPU: Graphics Processing Unit

Overview

GPU kernel generation turns operations or workload specifications into device code that runs efficiently on accelerators, where data movement, parallelism, and hardware-specific details strongly affect performance. Approaches include compiler and domain-specific language (DSL) systems with auto-scheduling, as well as LLM-based agents that generate CUDA or Triton code and refine it using profiling, execution feedback, reinforcement learning, or evolutionary search. Verification and benchmarking are central: a kernel must preserve numerical behavior across shapes and inputs while delivering speedups on target hardware. Systems can outperform framework baselines on selected tasks, but matching expert libraries and generalizing across hardware and production workloads remain difficult, and kernel-level gains do not always improve end-to-end runtime.

Latest papers

CardsList

74 papers7 papers in the last four weeks, up 75% on the four weeks before. 0.1% of all new papers.

  1. KernelMosaic: An Open Dataset and Agentic Framework for CUDA Kernel Generation

    Nov 10, 2026Ningyuan Xi, Yuhang Lai, Ning MiaoGPU Kernel GenerationGPU Kernel Optimization

  2. Correct First, Fast Second: What a Local 7B Code Model Can and Cannot Do for GPU Kernel Optimization

    Nov 9, 2026Jui-Chien Tsou, Po-Yi WuGPU Kernel GenerationLarge Language Model-Guided Optimization

  3. Ask, Don't Equip: Question-Driven Coding Agents for Deep Kernel Optimization

    Nov 2, 2026Gabriele Oliaro, Jaeseong Lee, Zhaoyuan Su +3GPU Kernel OptimizationGPU Kernel Generation

  4. LARK: Lean-Assisted Reasoning for LLM-Generated GPU Kernels

    Nov 1, 2026Zixi Zhang, Jiayi Nie, Haoran Wu +7GPU Kernel Correctness VerificationGPU Kernel Generation

  5. God Does Not Play Dice: Enforcing Batch Invariance in Agentic Kernel Generation

    Oct 27, 2026Youwei Bao, Chengpeng Wang, Simin ChenGPU Kernel GenerationGPU Kernel Optimization

  6. D2K-Bench: Can LLM Agents Turn Expert Designs into Efficient GPU Kernels?

    Oct 2, 2026Daifeng Li, Huiqiang Jiang, Chengruidong Zhang +6GPU Kernel GenerationGPU Kernel Optimization

  7. AI as a Compiler: Compiling Triton kernels without the Triton compiler

    Sep 29, 2026François Costa, Charly Castes, Thomas Bourgeat +1GPU Kernel OptimizationLarge Language Model-Guided Optimization

  8. KernelZero: Co-Evolving Proposer and Coder for Continuously Improved GPU Kernel Generation

    Sep 27, 2026Changxin Ke, Rui Zhang, Zixiang Fang +6GPU Kernel GenerationLLM-Based Program Synthesis

  9. Measuring the Checker: Mutation Analysis for GPU-Kernel Benchmark Oracles

    Sep 2, 2026Mingzhe Du, Anh Tuan Luu, Dong Huang +1GPU Kernel Correctness VerificationBenchmark Auditing

  10. DataKernelBench: Can LLMs Optimize Database Queries on GPUs?

    Aug 25, 2026Gokul Karthik Kumar, Yotam Perlitz, Corey Lammie +2GPU Kernel OptimizationGPU Kernel Generation

  11. HIERA: Workload-Aware Planning Across Implementation Spaces for GPU Kernel Optimization

    Aug 21, 2026Jinghao Wang, Qiqi Gu, Chenpeng Wu +3GPU Kernel OptimizationGPU Kernel Generation

  12. PTXBench: Benchmarking and Adapting LLMs for GPU Kernel Optimization with Architecture-specific PTX

    Aug 18, 2026Genghan Zhang, Yixin Dong, Chengze Fan +4LLM EvaluationGPU Kernel Optimization

  13. KernelArc: A Multi-Agent Framework for GPU Kernel Optimization

    Aug 17, 2026Joyjit Kundu, Ben Stoffelen, Kaili Wang +2GPU Kernel OptimizationGPU Kernel Generation

  14. CAKE: Compiler-Agent Co-Design for Frontier Kernel Evolution

    Aug 12, 2026Zihao Ye, Yingyi Huang, Hongyi Jin +11GPU Kernel OptimizationGPU Kernel Generation

  15. RealisticTritonBench: A Benchmark for Triton-Kernel Generation in Real-World AI Frameworks

    Aug 12, 2026Jinjun Huang, Zhongzhen Wen, Tongtong Xu +3LLM EvaluationGPU Kernel Generation

  16. SparseDitto: An Agentic Sparse Compilation Framework through Architecture-Aware Synthesis on GPUs

    Aug 5, 2026Shiyang Li, Guangyan Sun, Jinwei Tang +3GPU Kernel OptimizationHigh-Performance Computing

  17. Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Accelerator Kernels

    Aug 3, 2026Yansong Sun, Shenxiu Wu, Siyuan Chen +6Automated Program RepairGPU Kernel Generation

  18. Multi-turn RL with Structural and Performance Aware Rewards for CUDA Kernel Generation

    Jul 23, 2026Quazi Ishtiaque Mahmud, Nesreen K. Ahmed, Ali JannesariRL for Code GenerationGPU Kernel Optimization

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

    Jul 22, 2026Peiyu Zang, Jian Tao, Jialing Zhang +4GPU Kernel OptimizationBenchmark Design