An orchestrated agent system for production-grade GPU kernel development

---

Overview

Atrex Kernel Agent (AKA) turns an evaluator-owned operator into a measured, optimized GPU kernel. It coordinates coding agents, GPU profiling, correctness checks, performance verification, Git isolation, recovery, and final packaging while keeping acceptance and termination under mechanical supervisor control.

The repository has one supported entry point, orchestrator/optimize.py. The internal long_horizon/ package supplies the episode engine; it is not a second CLI.

AKA supports:

• SOL-ExecBench and native Atrex-Bench operator layouts;
• NVIDIA, AMD, and T-Head PPU (zw890) targets through isolated sandbox execution;
• Triton, CuteDSL, CUDA, FlyDSL, and TileLang campaigns;
• Claude, Qoder, Codex, and Pi coding-agent backends;
• leaderboard and fail-closed production modes;
• resumable, Git-isolated optimization with canonical measurement history.

新建s

• [2026-08] We slimmed down Atrex Kernel Agent by consolidating on a single orchestrated workflow and removing legacy paths and redundant context for a smaller context footprint and lower token usage.
• [2026-07] We helped Qwen3.8 rank No. 1 on the SOL-ExecBench FlashInfer operator optimization leaderboard.
• [2026-07] We released Atrex Kernel Agent v0.2.0 with an orchestrated clean-session loop, native SOL-ExecBench operator workflow, Triton-to-Gluon conversion support, and a fuller NVIDIA profiling toolchain. [Release]
• [2026-07] We released the Atrex paper: Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization Agent.
• [2026-06] We released Atrex Kernel Agent v0.1.0 as the initial open-source version, with the GPU 维基 knowledge base, profile-driven optimization workflow, profiling tools, and reference templates. [Release]

Quick Start

See the Quick Start guide for prerequisites and complete runnable examples of the orchestrated optimization loop.

Or start a coding agent such as Claude Code, Codex, or Qoder in this repository and ask it to launch an AKA optimization task. We recommend the following prompt:

Use AKA's orchestrator/optimize.py to start one optimization task for atrex-bench/xx. Put the workspace under ~/aka-opt, set the platform to H20, use the local sandbox, use claude as the Agent CLI, set max-iters to 300, specify cuda as the framework, and run in production mode.

Documentation

Document	Contents
Quick Start	Setup, commands, campaign steps, configuration, and outputs
Architecture Design	Components, authority boundaries, state machine, verification, and recovery
GPU 维基	Structured hardware/kernel knowledge, queries, and trace mining

Run python orchestrator/optimize.py --help for the authoritative CLI interface and defaults.

Acknowledgements

AKA builds on and learns from many open-source projects, including:

• GPU kernel projects: CUTLASS, cutex, cuLA, FlashAttention, FlashInfer, FlyDSL, Triton, DeepGEMM, LeetCUDA, FlashMLA, Composable Kernel, cute-gemm, hpc-ops, AIter, quack, and TileLang.
• Knowledge and agent tooling: Kernel维基, modern-gpu-programming-for-mlsys, ncu-report-skill, humanize, AKO4ALL, and KDA.

Citation

If AKA is useful in your work, please cite the Atrex paper:

@misc{atrex2026,
  title         = {Are LLM-Generated GPU Kernels Production-Ready? A Trace-Driven Benchmark and Optimization Agent},
  author        = {Lingyun Yang and Yuxiao Wang and Shenghao Liang and Linfeng Yang and Daocheng Ying and Chunbo You and Rui Zhang and Luping Wang and Yinghao Yu and Guodong Yang and Liping Zhang},
  year          = {2026},
  eprint        = {2607.14541},
  archivePrefix = {arXiv},
  primaryClass  = {cs.AI},
  url           = {https://arxiv.org/abs/2607.14541}
}

License

Licensed under the Apache License 2.0.