Discover, run, and compose AI agents from any framework

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

Your friendliest open source AI automation tool ✨ Workflow automation tool 200+ integration / Enterprise automation tool / Zapier Alternative

🪄 Create rich visualizations with AI

deepseek running locally in your browser

Python tool for converting files and office documents to Markdown.

Composable building blocks to build Llama Apps

A pure Rust implementation of WebRTC

Everything about the SmolLM2 and SmolVLM family of models

A fast multimodal LLM for real-time voice

Open-source developer platform to power your entire infra and turn scripts into webhooks, workflows and UIs. Fastest workflow engine (13x vs Airflow). Open-source alternative to Retool and Temporal.

first base model for full-duplex conversational audio

Verbatim Automatic Speech Recognition with improved word-level timestamps and filler detection

Home of OpenVMM and OpenHCL.

Official code for "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"

[ICLR 2025] Agent S: an open agentic framework that uses computers like a human

Implementing the 4 agentic patterns from scratch

Making the community's best AI chat models available to everyone.

⚡ The future of programmable SIP servers.

An extremely fast implementation of whisper optimized for Apple Silicon using MLX.

A curated list of awesome JSON Schema resources, tutorials, tools, and more

📖 A step-by-step guide for beginners to running Claude-generated React code locally.

a super simple, no-nonsense diagramming library written in react that just works

async fn(Request) -> Result<Response, Error>

A native gRPC client & server implementation with async/await support.

VideoSys: An easy and efficient system for video generation

RobustMQ is a next-generation, high-performance, cloud-native, converged message queue that is compatible with multiple mainstream message queuing protocols and has complete Serveless capabilities.

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.

open-source multimodal large language model that can hear, talk while thinking. Featuring real-time end-to-end speech input and streaming audio output conversational capabilities.