Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.
-
Updated
Nov 13, 2026 - Python
Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.
Hands-on LLM algorithms and systems with PyTorch, Triton & CUDA.《大模型算法与系统教程》:模型架构与训练、后训练、推理加速、性能与算子优化,配套 Notebook 练习、答案解析、测试与实验。
Fastest MoE/LLM inference runtime for consumer and edge Blackwell GPUs. SN74 on Gittensor.
Open source skill library for AI coding agents to write, optimize, and debug high performance compute kernels across CUDA, Triton, and quantized workloads.
A lightweight, general-purpose framework for evaluating GPU kernel and benchmark.
Lightweight desktop tool for Call of Duty: Modern Warfare 4. Smart FPS optimization, performance tuning, process management and real-time monitoring. Designed to improve frame rates and stability. Regularly updated.
Pure C++ and HIP inference for gpt-oss on AMD GPUs - no rocBLAS, no RCCL, no MPI. Every kernel, collective and the tokenizer written from scratch. 84,451 tok/s on the 20B and 40,271 on the 120B, on one node of 8x MI250.
Custom Linux kernels purpose-built for Apple Mac hardware
🔥 The living map of LLM-driven kernel generation: 280+ agents, models, benchmarks, and industry systems across NVIDIA, AMD, Ascend, TPU, and more. Continuously updated.
Repo containing artifacts for Neurips 2025 tutorial- How to Build Agents to Generate Kernels for Faster LLMs (and Other Models!)
可验证的 CUDA 学习主线:SGEMM、通用 GPU 算子、性能优化与轻量推理组件
Agent-queryable ROCm kernel optimization knowledge base for AMD Instinct MI300/gfx942 and MI350/MI355X/gfx950, packaged for Codex CLI and Claude Code with merged-PR provenance, real-silicon validation, and a maintainer-controlled pull-request evidence pipeline.
Open reproduction of NVIDIA's AVO paper (arXiv:2603.24517): evolutionary search where an autonomous coding agent IS the variation operator — Vary(P)=Agent(P,K,f). Runs on the Claude Code or Codex session you already have.
Evidence-driven CUDA, CUTLASS, Triton and GPU workload optimization for ChatGPT · 使用 ChatGPT 驱动 GPU workload 性能优化
Portable Ascend C and TileLang operator engineering skill with tested workflows and lessons from DeepGEMM and TileKernels.
METAL-SCI: a scientific compute benchmark for evolutionary LLM kernel search on Apple Silicon Metal
RWKV-7 FP8 quantized inference - 6.4x decode speedup on Blackwell GPUs with <0.3% accuracy loss. Full FP8 E4M3 weight quantization with fused Triton kernels.
Fast Triton kernels for multi-scale deformable attention (MSDA) — the core operator behind Deformable DETR, DINO, and Mask2Former — plus a from-scratch course and hands-on labs
Open-source H200 CUDA kernels for the Muse Spark Open Kernel Challenge (OKC): decode layer 2.21x, NVFP4 MoE 2.40x, BF16 proj 1.55x, prefill 1.82x, attention 2.65x vs SGLang; judge results, measured H200 facts, negative results
To associate your repository with the kernel-optimization topic, visit your repo's landing page and select "manage topics."