L'accès à l'art et au divertissement ne devrait jamais être un luxe ni un parcours du combattant.

A Foundation Model for Generalist Gaming Agents

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

A browser-based desktop where AI Agent operates every app through natural language.

An open-source Collaborative Multi-Agent OS for transparent, human-in-the-loop task coordination via Matrix rooms.

A local AI companion with memory, emotion, voice, avatar control, and agentic features — built for real-time interaction on low VRAM.

Official repository of Utonia: Toward One Encoder for All Point Clouds

Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.

Virtual whiteboard for sketching hand-drawn like diagrams

The open-source voice synthesis studio powered by Qwen3-TTS.

The best ChatGPT that $100 can buy.

Moshi is a speech-text foundation model and full-duplex spoken dialogue framework. It uses Mimi, a state-of-the-art streaming neural audio codec.

Easy install, Ultra-lightweight secure AI assistant. Inspired by OpenClaw. 🦀

Fun-Audio-Chat is a Large Audio Language Model built for natural, low-latency voice interactions.

Soprano: Instant, Ultra-Realistic Text-to-Speech

Open-source text-to-speech for European languages with voice cloning

A complete Spotify clone built with Typescript, React, React Redux, Spotify Web API, and Spotify Playback SDK. This web client replicates the core functionalities of Spotify, including music playba…

Local voice assistant that learns new abilities via auto-discovered n8n workflows exposed as tools via MCP

PersonaLive! : Expressive Portrait Image Animation for Live Streaming

experiments in goose

an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM

Faster Whisper transcription with CTranslate2

A tweak to enhance Spotify experience

vllama is an open source hybrid server that combines Ollama's seamless model management with vLLM's lightning-fast GPU inference, delivering a drop-in OpenAI-compatible API for optimized performance.

Native Multimodal Models are World Learners

YouTube app for webOS without ads