Open source. Open possibilities.

Tabby is a self-hosted, open-source AI coding assistant designed as an on-premises alternative to GitHub Copilot.

A curated collection of 90+ hands-on AI engineering tutorials and projects covering LLMs, RAG, agents, MCP, multimodal apps, fine-tuning and evaluation, organized by beginner, intermediate and advanced difficulty.

Made With ML is an open-source course and codebase teaching how to design, develop, deploy and iterate on production-grade machine learning applications, covering MLOps, testing, serving, CI/CD and continual learning.

GPT Researcher is an open-source autonomous agent that performs deep research using web and local documents, generating detailed reports with citations. It supports multi-agent workflows, image generation, and integrates with Claude via MCP.

Open-source collection of 100+ AI agents, agent skills, and RAG apps. Apache-2.0, hand-built and tested end-to-end, works with Claude, GPT, Gemini, DeepSeek, Llama, Qwen.

Open WebUI is a self-hosted, extensible AI platform that provides a user-friendly interface for local and cloud-based models, supporting Ollama and OpenAI-compatible APIs.

QwenPaw is a self-hosted personal AI assistant that runs locally or in the cloud, with three-layer memory, sandboxed tool execution, multi-agent support, and connectors for DingTalk, Lark, WeChat, Discord, Telegram and more.

nanobot is an ultra-lightweight, self-hosted personal AI agent framework written in Python, featuring a WebUI, terminal interface, and integrations with various chat applications.

Micro Opinion (BettaFish) is a multi-agent public opinion analysis system built from scratch, with AI crawler clusters, multimodal analysis, forum collaboration, and intelligent reports to break information bubbles, restore public sentiment, predict trends, and assist decisions. Supports 30+ social platforms, private database access, pure Python, Docker deployment.

YuE2 is an open music generation model that unifies symbolic planning with audio synthesis: it turns lyrics and a style prompt into an editable melody-and-chord score, then renders a full song with vocals and accompaniment. It also supports zero-shot covers and agentic score editing.

Unsloth is a desktop application and framework for running and training large language models (LLMs) and diffusion models locally across Windows, macOS, and Linux.

A text-based 5e-compatible RPG where an AI Dungeon Master uses an MCP/SQLite game engine for real dice rolls, persistent stats, spell slots, and combat.

AutoAgent is a zero-code framework that lets users create and deploy LLM agents and workflows using natural language alone, with features like automated agent profiling, tool creation, and multi-agent orchestration.

MiroFish is a multi-agent swarm intelligence prediction engine that simulates social dynamics from seed data. Users provide context (news, scenarios), and thousands of LLM-powered agents with memory and personas interact in a digital sandbox to generate prediction reports and interactive simulations.

A self-hosted web UI for AI coding agents: Rust backend (axum, tokio, SQLite) with an embedded Flutter web frontend, pluggable providers for Devin CLI, OpenCode and Codex CLI, browser login, sessions and file handling.

Oumi is an open-source platform for the full foundation-model lifecycle: data preparation, training and fine-tuning (SFT, LoRA, QLoRA, GRPO), evaluation, data synthesis with LLM judges, and deployment via vLLM/SGLang, with a CLI and cloud job launching.

FastMCP is a Python framework for building Model Context Protocol (MCP) applications, enabling seamless integration of LLMs with tools and data via servers, clients, and interactive UIs.

UL-SMF is a KV cache compression framework for long-context Transformer inference, claiming up to 384x compression via FSQ and 16D latent mapping. The core compression logic requires a proprietary closed-source binary.

A modular graph-based Retrieval-Augmented Generation (RAG) system from Microsoft Research. It extracts structured knowledge graphs from unstructured text using LLMs to improve question-answering over private data. Now in maintenance mode with bug fixes and dependency updates only.

Open-source Python framework for building production-grade multi-agent AI workflows using role-based Crews and event-driven Flows. Combines autonomous agent collaboration with precise control, supporting tools, memory, structured outputs, MCP/A2A, and multiple LLM connections.