TokenSpeed is a speed-of-light LLM inference engine.

High Performance LLM Inference Operator Library

A Triton JIT runtime and ffi provider in C++

Efficient Triton Kernels for LLM Training

Towards Holistic evaluation of Generative Diffusion Transformers!

An LLM post-training framework with vLLM for RL Scaling

Skills for writing tilelang and debugging with CUDA toolkits.

Open Neural Network Exchange to C compiler.

Large Language Model (LLM) Systems Paper List

从零开始玩转OpenClaw：最全面的中文教程，涵盖安装、配置、实战案例和避坑指南（github版）

claude code profiler

TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)

let coding agents use ncu skills analysis cuda program automatically!

Glamourous agentic coding for all 💘

[ICLR2025 Spotlight] SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models

Fast and memory efficient c++ flat hash table/map/set

Claude Code agent skill for autonomous AI/scientific research workflows

Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepowe…

KsanaDiT: High-Performance DiT (Diffusion Transformer) Inference Framework for Video & Image Generation

from vibe coding to agentic engineering - practice makes claude perfect

Material for gpu-mode lectures

how to optimize some algorithm in cuda.

AI Tensor Engine for ROCm

A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs dir…

LLM Inference via Triton (Flexible & Modular): Focused on Kernel Optimization using CUBIN binaries, Starting from gpt-oss Model

Create beautiful slides on the web using a coding agent's frontend skills