awesome-opensource-ai

Flexible, lightweight open-source framework for orchestrating multiple AI agents to handle complex conversations with parallel execution capabilities. Apache 2.0 licensed.

Multi-agent orchestration runtime with Kanban-style team management, message review loops, and provider integrations for reusable agentic teams.

Agent2Agent (A2A) open protocol enabling communication and interoperability between opaque agentic applications. Donated to Linux Foundation by Google with 50+ technology partners. Apache 2.0 licensed.

Open-source automation platform with AI agents, MCP integrations, and self-hosted workflow orchestration.

Library to build and auto-optimize LLM applications with LLM-AutoDiff for fine-tuning-free optimization. End-to-end workflow optimization with tracing and human-in-the-loop capabilities. MIT licensed.

Reliable multi-agent orchestration framework built on top of the OpenAI Assistants API with organizational structure modeling.

Local-first music generation model with broad hardware support across Mac, AMD, Intel, and CUDA devices.

Comprehensive toolkit for Audio, Music, and Speech Generation (9.7K stars).

Open-source LLMOps platform combining prompt playground, prompt management, LLM evaluation, and observability.

Self-hosted ML experiment tracker designed to handle 10,000s of training runs with performant UI and SDK for programmatic access. Apache 2.0 licensed.

Kubernetes resource scheduler for batch, data, and ML workloads. Provides hierarchical resource queues, multi-tenancy fairness, and gang scheduling for big data and machine learning applications. Apache 2.0 licensed.

CNCF graduated container-native workflow engine for orchestrating parallel jobs on Kubernetes. Powers Kubeflow Pipelines and widely used for ML/data processing at scale. Apache 2.0 licensed.

Unified framework to build, ship, and scale AI apps.

Collection of deep learning models and utilities optimized for Cerebras hardware. Includes reference implementations for Llama, Mixtral, DINOv2, and Llava with configuration files, data preprocessing tools, and checkpoint converters. 1,150+ stars. Apache 2.0 licensed.

Comprehensive benchmark to evaluate LLMs as agents across 8 diverse environments including household, web shopping, OS interaction, and database tasks. ICLR 2024. Apache 2.0 licensed.

RAG AutoML tool for automatically finding optimal RAG pipelines. Evaluates and optimizes retrieval-augmented generation with AutoML-style automation for your own data and use-case. Apache 2.0 licensed.

Gym environment for web task automation and agent evaluation. Includes MiniWoB, WebArena, WorkArena, and more. Apache 2.0 licensed.

Web app for interacting with any LangGraph agent (Python & TypeScript) via a chat interface. Stream messages, handle interruptions, and view agent state. MIT licensed.

Self-hosted cyber companion and VTuber platform with Live2D character rendering, real-time voice chat, and game integration capabilities. MIT licensed.

All-in-one RAG + agents platform.

Multi-channel personal AI agent assistant and integration framework for messaging platforms, LLMs, and plugins. AGPL-3.0 licensed.

AI suite for power users with multi-model "Beam" chats, AI personas, voice, text-to-image, code execution, and PDF import. MIT licensed.

Self-hosted webUI for 30+ generative AI models. Generate multimedia content with AI on your own computer, even without dedicated GPU (8GB RAM minimum). Works offline once deployed. GPL-3.0 licensed.

Security and compliance toolkit for autonomous agents with policy enforcement, zero-trust identity, and sandboxed execution to reduce operational risk. MIT licensed.

Agentic LLM vulnerability scanner and AI red teaming kit with multi-step attack simulation and automated security probing. Apache 2.0 licensed.

Python SDK for AI agent monitoring, LLM cost tracking, benchmarking, and evaluation. Integrates with CrewAI, Agno, OpenAI Agents SDK, LangChain, Autogen, AG2, and CamelAI. MIT licensed.

Comprehensive toolkit for detecting, understanding, and mitigating unwanted algorithmic bias in datasets and ML models.

Full-stack AI Red Teaming platform securing AI ecosystems via OpenClaw Security Scan, Agent Scan, Skills Scan, MCP scan, AI Infra scan and LLM jailbreak evaluation. Apache 2.0 licensed.

Comprehensive AI explainability toolkit with interpretability algorithms for data and machine learning models. Includes TED, BRCG, and ProtoNN methods for diverse explanation needs. Apache 2.0 licensed.

Local-first session intelligence and cost analytics dashboard for AI coding agents, supporting Claude Code, Codex, and other tools. MIT licensed.

Official AWS-supported MCP servers, skills, and plugins to help AI coding agents build, deploy, and manage applications on AWS. Apache 2.0 licensed.

Local-first TUI for observing AI coding agent sessions across Claude Code, Codex CLI, Gemini CLI, Aider, Cursor exports, OpenCode, and more.

All-in-one LLM CLI in Rust featuring Shell Assistant, Chat-REPL, RAG, AI Tools & Agents. Supports 20+ providers. MIT/Apache 2.0 licensed.

CLI that writes your Git commit messages for you with AI. Never write a commit message again. Supports multiple providers including OpenAI, Groq, xAI, Ollama, and LM Studio. MIT licensed.

Terminal-based AI pair programmer. Edit code in your local editor and aider implements the changes. Supports multiple LLMs, voice coding, and automatic Git commits. Top scores on SWE Bench. Apache 2.0 licensed.

Open-source textbook and code repository covering AI agent design principles, architectures, and engineering practice. Apache 2.0 licensed.

Comprehensive curriculum covering machine learning, deep learning, NLP, computer vision, and agents by implementing them from scratch. MIT licensed.

93+ production-ready projects with in-depth tutorials on LLMs, RAG, and real-world AI agent applications. Comprehensive resources for all skill levels from beginner to advanced. MIT licensed.

12-week, 24-lesson curriculum on Artificial Intelligence. Covers symbolic AI, neural networks, computer vision, NLP, and reinforcement learning with hands-on labs.

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls. Principles: Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution. MIT licensed.

Curated list of artificial intelligence courses, books, video lectures, and papers for developers and researchers. MIT licensed.