Blazingly fast LLM inference.

The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, No-code agent builder, and more.

Tool to convert Go Templates to Jinja2 Templates

A single file multi-platform (Unix, Windows) C++ header-only linenoise-based readline library.

InstructLab Core package. Use this to chat with a model and execute the InstructLab workflow to train a model using custom taxonomy data.

ROCm Application for Reporting System Info

Blazingly fast terminal recorder that generates animated gif images for the web written in rust

A minimalistic C++ Jinja templating engine for LLM chat templates

Launch configurable virtual machines with libkrun

Experimental KVM-based type-2 hypervisor in Rust implemented from scratch.

Get your documents ready for gen AI

Tensor library for machine learning

User-friendly AI Interface (Supports Ollama, OpenAI API, ...)

Stable Diffusion and Flux in pure C/C++

The next generation Linux workstation, designed for reliability, performance, and sustainability.

uBlock Origin - An efficient blocker for Chromium and Firefox. Fast and lean.

Source code behind the Windows Subsystem for Linux documentation.

Awesome list dedicated to Windows Subsystem for Linux

Concatenate a directory full of files into a single prompt for use with LLMs

OCI Artifact for ML model & metadata

Python bindings for llama.cpp

LLM inference in C/C++

The goal of RamaLama is to make working with AI boring.