Stars
[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
DeepSeek 4 Flash and PRO local inference engine for Metal, CUDA and ROCm
An open source card-based battle engine written in modern C++
Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels
Reverse engineering NVIDIA SASS instruction dictionary, kernel audits and pattern recognition across GPU architectures.
A lightweight image editor for Windows, MacOS, Linux with a focus on intuitive use and modern UI design. Features layers, filters, brushes, selection tools, color adjustments, animated images suppo…
A pure Python HTML5 parser that just works. No C extensions to compile. No system dependencies to install. No complex API to learn.
Library for reducing tail latency in RAM reads
A CMake build platform and dependency provider with reusable CI (continuous integration) pipelines.
GEMMul8 (GEMMulate): GEMM emulation and its extension to BLAS-like matrix operations using INT8/FP8 matrix engines based on the Ozaki Scheme II
A complete guide on getting Nvidia's Proprietary drivers for Fedora desktops (Workstation, KDE and Atomic) with instructions for Secure Boot and encrypted drives.
IntelliJ plugin that helps developing Ansible files
Public repository of the Micro QuickJS Javascript Engine
An extremely fast Python linter and code formatter, written in Rust.
An open-source application for biological image analysis
Next generation Cosmic desktop environment
MuleSoft AI Chain Connector
A friendly programming language from the future
Python-based continuous integration testing framework; your pull requests are more than welcome!
Large Language Model (LLM) Systems Paper List
a language for fast, portable data-parallel computation
A next generation Python CMake adaptor and Python API for plugins
Arbitrary precision integer and rational arithmetic library