Laurent Payot

colibri. Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

tura. Across 348 long-horizon benchmark sessions, Tura used up to 83.1% fewer turns on the rewrite benchmark and improved the DeepSWE pass rate by up to 16.7 percentage points compared with Codex CLI.

headroom. Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

codegraph. Pre-indexed code knowledge graph, auto syncs on code changes, for Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, and Hermes Agent — fewer tokens, fewer tool calls, 100% local

rtk. CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies

oh-my-pi. ⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more

pi-hooks. Claude Code-compatible command hooks for the Pi coding agent

myna. Myna is a lightweight speech-to-text application for Ubuntu Desktop.

superpowers. An agentic skills framework & software development methodology that works.

agent-skills. Production-grade engineering skills for AI coding agents.

lancedb. Developer-friendly OSS embedded retrieval library for multimodal AI. Search More; Manage Less.

sky. Sky — an Elm-inspired language that compiles to Go. Hindley-Milner types, server-driven UI (Sky.Live), single binary output.

CWR. Arma: Cold War Assault Remastered Source Code Repository.

ant. javascript for 🐜's, a tiny runtime with big ambitions

Bonsai-demo. Bonsai Demo

cavemem. Cross-agent persistent memory for coding assistants. Stored compressed. Retrieved fast. Local by default.

chrome-devtools-mcp. Chrome DevTools for coding agents

DFlash_Qwen3.6_27B_LlamaCPP. Python

llama-ai. A project repository for work on improving local LLMs on my personal AMD devices

CachyLLama. LLM inference in C/C++

CLIO. An AI-assisted coding agent that runs in your terminal and supports many providers and models.

kage. Shadow any website for offline viewing, with the JavaScript stripped out

mgrep. A calm, CLI-native way to semantically grep everything, like code, images, pdfs and more.

merman. Mermaid.js, but headless, in Rust.

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

ECC. The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

vm0. Zero, your trustworthy AI teammate for real work.

knowledge-work-plugins. Open source repository of plugins primarily intended for knowledge workers to use in Claude Cowork

adhd. ADHD — a skill for coding agents. Tree-of-thought with pruning, built on the Claude & Codex Agent SDK. Fans out parallel divergent thoughts under different cognitive frames, scores, prunes traps, deepens the survivors. The no-brainer skill for creative and interdisciplinary work.

s-stack. Scott's Dope Skills

tts-bench. Speed and samples benchmark: for all types of text to speech (TTS) models on Windows/Linux/Mac.

MiMo-Code. MiMo Code: Where Models and Agents Co-Evolve

agentmail-skills. Python

KittenTTS. State-of-the-art TTS model under 25MB 😻

reference-sdks. A curated collection of workshop SDKs.

Gentle-Coding. An ongoing, collaborative meta-analysis about Human-AI-Interactions. We aggregate data and knowledge to build a non-abrasive, user-friendly prompting framework tailored to LLM mechanics, ensuring reasoning stability and a friction-free prompting environment that is safe for the human psyche and wellbeing.

amd-strix-halo-toolboxes. Python

TheRock. The HIP Environment and ROCm Kit - A lightweight open source build system for HIP and ROCm

nanoclaw. 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 directly on Anthropic's Agents SDK

little-coder. A harness optimized to smaller LLMs

smallcode. AI coding agent optimized for small LLMs. 87% benchmark with 4B-active model.

opencode-claude-auth. OpenCode plugin that uses your existing Claude Code credentials — no separate login needed.

nanostores. A tiny (340 bytes) state manager for React/RN/Preact/Vue/Svelte with many atomic tree-shakable stores

skills. Skills for Real Engineers. Straight from my .agents directory.

designer-skills. A collection of agent skills for designers who prototype and build with AI coding tools. These skills encode design process so AI follows a structured path instead of producing random output.

lucidrag. A full RAG (Retrieval Augmented Generation) system featuring a low friction GraphRAG and conversational interface on documents.

honcho. Memory library for building stateful agents

browser-harness. Browser Harness | Self-healing harness that enables LLMs to complete any task.

agent-browser. Browser automation CLI for AI agents

caveman. 🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

beads. Beads - A memory upgrade for your coding agent

dolt. Dolt – Git for Data

wallace-cli. Pretty CSS analytics on the CLI

fallow. Codebase intelligence for TypeScript and JavaScript. Free static analysis of code and styles: unused code, duplication, circular deps, complexity hotspots, architecture boundaries, design-system drift. Optional paid runtime layer (Fallow Runtime): hot-path review and cold-path deletion evidence from real production traffic.

dagger. Automation engine to build, test and ship any codebase. Runs locally, in CI, or directly in the cloud

dflash. DFlash: Block Diffusion for Flash Speculative Decoding

llama-swap. Reliable model swapping for any local OpenAI/Anthropic compatible server - llama.cpp, vllm, etc

moonshine. Very low latency speech to text, intent recognition, and text to speech, for building voice agents and interfaces

llms-txt-hub. 🤖 The largest directory for AI-ready documentation and tools implementing the proposed llms.txt standard

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

anything-llm. Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience

evo-x2-power-mode. Rust

hermes-agent. The agent that grows with you

llama.cpp. LLM inference in C/C++

PrismML-Bonsai-vs-Qwen3.5-Benchmark. PrismMl Bonsai vs Qwen3.5 Benchmark

pi-updater. Auto updater for pi

pi-dcp. TypeScript

llm-knowledge-cutoff-dates. This repository contains a summary of knowledge cut-off dates for various large language models (LLMs), such as GPT, Claude, Gemini, Llama, and more.

sqlite-vec. A vector search SQLite extension that runs anywhere!

pi-lens. Real-time code feedback for pi — LSP, linters, formatters, structural analysis

vite-plus. Vite+ is the unified toolchain and entry point for web development. It manages your runtime, package manager, and frontend toolchain in one place.

paraglide-js. Compiler-based i18n library that emits tree-shakable translations, leading to up to 70% smaller bundle sizes.

clvk. Implementation of OpenCL 3.0 on Vulkan

pi-autoresearch. Autonomous experiment loop extension for pi

agents.md. AGENTS.md — a simple, open format for guiding coding agents

FastFlowLM. Run LLMs on AMD Ryzen™ AI NPUs in minutes. Just like Ollama - but purpose-built and deeply optimized for the AMD NPUs.

playwright-mcp. Playwright MCP server

pi-mcp-adapter. Token-efficient MCP adapter for Pi coding agent

BrowserOS. 🌐 The open-source Agentic browser; alternative to ChatGPT Atlas, Perplexity Comet, Dia.

aie-book. [WIP] Resources for AI engineers. Also contains supporting materials for the book AI Engineering (Chip Huyen, 2025)

vocalinux. Free, open-source, 100% offline voice dictation for Linux. Speak and type anywhere via whisper.cpp, Whisper & VOSK engines, GPU-accelerated, works on X11 + Wayland!

claw-eval. Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.

lemonade. Lemonade helps users discover and run local AI apps by serving optimized LLMs right from their own GPUs and NPUs. Join our discord: https://discord.gg/5xXzkMu8Zk

qwen-code. An open-source AI coding agent that lives in your terminal.

kon. Kon is a minimal coding agent (and also a highly opinionated one)

pi. AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI

youtube-tutorials. All my YouTube tutorials. Everything here is free - but please follow me on YouTube if you find these helpful. ❤️

rescript-nodejs. Node bindings for ReScript

ydotool. Generic command-line automation tool

whisper.cpp. Port of OpenAI's Whisper model in C/C++

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

voxd. VOXD is a speech-to-text, voice-typing, dictation software for linux distributions. It is an open-source, free of charge, USER-FRIENDLY software, for as many linux distros as possible.

dsnote. Speech Note Linux app. Note taking, reading and translating with offline Speech to Text, Text to Speech and Machine translation.

opencode. The open source coding agent.

ungoogled-chromium. Google Chromium, sans integration with Google

GLM-5. GLM-5: From Vibe Coding to Agentic Engineering

koboldcpp. Run GGUF models easily with a KoboldAI UI. One File. Zero Install.

rescript-react-navigation. ReScript bindings for React Navigation

rescript-react-native. ReScript bindings for React Native

zed-mermaid. Mermaid support for Zed

rescript-zed. ReScript support for Zed editor

Step-3.5-Flash. Fast, Sharp & Reliable Agentic Intelligence

sury. 🧬 The fastest schema with next-gen DX

oxc-zed. Oxc extension for Zed

mistral-vibe. Minimal CLI coding agent by Mistral

free-llm-api-resources. A list of free LLM inference resources accessible via API.

better-comments-next. TypeScript

experimental-rescript-webapi. Experimental successor to rescript-webapi

vite-plugin-rescript. Seamlessly integrate ReScript with Vite

reanalyze. Experimental analyses for ReScript and OCaml: globally dead values/types, exception analysis, and termination analysis.

create-rescript-app. npm create rescript-app@latest

rescript-vscode. Official VSCode plugin for ReScript

rescript-react. Official ReScript bindings for ReactJS

Brahma.FSharp. F# quotation to OpenCL translator and respective runtime to utilize GPGPUs in F# applications.

brimstone. New JavaScript engine written in Rust

only-allow. Force a specific package manager to be used on a project

magentic-ui. MagenticLite is an experimental agent that works across the browser and local file system

Fable.Multiverse. A template to get up and running as fast as possible with Fable and publishing to multiple languages.

preact-signal-browser-router. A barebone hash and url router for browsers using preact+signals

wouter. 🥢 A minimalist-friendly ~2.2KB routing for React and Preact

preact-iso. Isomorphic utilities for Preact

Vectual. Open source charting library written in Elm

rolldown-vite. The WIP version of Vite powered by Rolldown

Fidelity.CloudEdge. Toolkit for building modern Cloudflare Solutions with F# and Fable Compiler

GeForce-Infinity. GeForce Infinity is a next-gen application designed to enhance the GeForce NOW experience. Originally created to address the lack of native GeForce NOW support on Linux, this app also provides functionality for macOS and Windows users. Our goal is to refine the user interface and offer custom features for an improved experience.

playwright. Playwright fork that works with Cloudflare Browser Rendering

Expecto.Template. .net core template for an Expecto (https://github.com/haf/expecto)

playwright-dotnet. .NET version of the Playwright testing and automation library.

Runfs. Prototype for "dotnet run app.fs"-like functionality

cloudflare-typescript. The official TypeScript library for the Cloudflare API

fp-filters. A curated list of ready-to-use (functional programming) array filters (TS / ESM / CJS)

ripple. the elegant TypeScript UI framework

languages. Compare languages

date-fns. ⏳ Modern JavaScript date utility library ⌛️

Perla. A cross-platform web standards-first SPA dev server

JoyShockMapper. A tool for PC gamers to play games with DualShock 4s, JoyCons, and Pro Controllers. Gyro aiming, flick stick.

datastar. The hypermedia framework.

material. material is a library of web components that helps build beautiful and accessible web applications. It uses Material 3, the latest version of Google's open-source design system.

material-web. Material Design Web Components - Community Edition

Qwen3-Coder. Qwen3-Coder is the code version of Qwen3, the large language model series developed by Qwen team.

gyromouse. A crossplatform mapper from gamepad inputs to keyboard and mouse actions, with special care for good gyro controls.

JoyShockMapper. A tool for PC gamers to play games with DualShock 4s, JoyCons, and Pro Controllers. Gyro aiming, flick stick.

elm-syntax-to-fsharp. transpile elm to F#

flips. Fsharp LInear Programming System

aitrack. Patched to work on linux!

opentrack. Head tracking software for MS Windows, Linux, and Apple OSX

aitrack. 6DoF Head tracking software

yt-dlp. A feature-rich command-line audio/video downloader

X1CarbonGen10. Keyboard with Trackpoint for Framework 16 laptop: hardware design

free-vscode-csharp. Free/Libre C# language support for VSCode and vscode-compatible editors

gallery. A gallery that showcases on-device ML/GenAI use cases and allows people to try and use models locally.

Feliz.DateRange. Feliz Binding for react-date-range

Proton. Compatibility tool for Steam Play based on Wine and additional components

Feliz.Router. A router component for React and Elmish that is focused, powerful and extremely easy to use.

uv. An extremely fast Python package and project manager, written in Rust.

LTX-Video. Official repository for LTX-Video

thinkeys. Split ortholinear custom replacement keyboard with TrackPoint for laptops

marmelade. Toy functional language in the ML spirit. With a bidirectional typer, Pratt-parsing and layout syntax over braces.

chain-of-draft. Code and data for the Chain-of-Draft (CoD) paper

MiMo. MiMo: Unlocking the Reasoning Potential of Language Model – From Pretraining to Posttraining

void. TypeScript

formation-fsharp. Supports de ma formation sur F# en français

hot_reload_poc. F#

Fumble. Thin F# API for Sqlite for easy data access to sqlite database with functional seasoning on top

Feliz.ChartJS. Feliz binding for React ChartJS.

FreeAct. F#

Fable.Pyxpecto. fable testing library targeting multiple languages

jan. Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.

SQLProvider. A general F# SQL database erasing type provider, supporting LINQ queries, schema exploration, individuals, CRUD operations and much more besides.

Html2Feliz. F#

ionide-analyzers. F#

maplibre-gl-js. MapLibre GL JS - Interactive vector tile maps in the browser

fd. A simple, fast and user-friendly alternative to 'find'

Mistral. C# SDK for the Mistral API -- chat completions, embeddings, and tool calling

typescript-go. Staging repo for development of native port of TypeScript

bluloco-zed. bluloco theme for zed editor

theme-bluloco-light. A fancy and sophisticated light color scheme.

syntax-highlighting-samples. A bunch of small code snippets for checking syntax highlighting when creating themes for editors.

FStar. A Proof-oriented Programming Language

monaspace. An innovative superfamily of fonts for code

Roo-Code. Roo Code gives you a whole dev team of AI agents in your code editor.

diku-canvas. Simple canvas library for F# (2d-drawing, loading/saving of image files, using SDL2)

open-deep-research. Open source alternative to Gemini Deep Research. Generate reports with AI based on search results.

sniffnet. Comfortably monitor your Internet traffic 🕵️‍♂️

Qwen3. Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

twinny. The most no-nonsense, locally or API-hosted AI code completion plugin for Visual Studio Code - like GitHub Copilot but 100% free.

DeepSeek-Coder-V2. DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

llama.cpp. LLM inference in C/C++

jscanify. The Javascript document scanning library.

ollama. Get up and running with Kimi-K2.6, GLM-5.2, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

vllm. A high-throughput and memory-efficient inference and serving engine for LLMs

DeepSeek-R1.

llama.vscode. VS Code extension for LLM-assisted code/text completion

fslang-suggestions. The place to make suggestions, discuss and vote on F# language and core library features

Fable.DateFunctions. Fable binding for date-fns javascript library, implemented as extension methods for DateTime. See https://zaid-ajaj.github.io/Fable.DateFunctions/

zed-fsharp. F# support for Zed

Fun.Blazor. Powered by .NET blazor!!! ❤ F#

smolagents. 🤗 smolagents: a barebones library for agents that think in code.

layx. An extensible CSS framework for modern web development. Combines modular CSS, customizable components, and a powerful CLI for seamless frontend and backend integration.

rolldown. Fast Rust bundler for JavaScript/TypeScript with Rollup-compatible API.