# Artificial Intelligence > Open-source AI models, agents, runtimes, gateways, and retrieval tooling. Navigation: [Index](../../llms.txt) · [HTML](https://awesomeopensource.dev/category/artificial-intelligence) ## 106 projects - [openclaw/openclaw](../openclaw-openclaw.md): Open-source AI assistant that lives on your own hardware and in your chat apps · Activity score 5.0/5 (Thriving) · TypeScript · ★ 394396 - [n8n-io/n8n](../n8n-io-n8n.md): Fair-code workflow automation with a visual canvas, custom code and 1500+ integrations · Activity score 5.0/5 (Thriving) · TypeScript · ★ 210408 - [langgenius/dify](../langgenius-dify.md): Visual platform for building LLM apps with RAG pipelines, agents and model integrations · Activity score 5.0/5 (Thriving) · Python · ★ 161115 - [open-webui/open-webui](../open-webui-open-webui.md): Self-hosted web UI for chatting with Ollama and OpenAI-compatible models offline · Activity score 5.0/5 (Thriving) · Svelte · ★ 158577 - [langchain-ai/langchain](../langchain-ai-langchain.md): Python framework for building agents and LLM applications from composable components · Activity score 5.0/5 (Thriving) · Python · ★ 149469 - [paperclipai/paperclip](../paperclipai-paperclip.md): Manage a team of AI agents with goals, budgets and task tracking · Activity score 5.0/5 (Thriving) · TypeScript · ★ 113685 - [debpalash/VoiceStudio](../debpalash-voicestudio.md): Local voice studio for cloning, dubbing, dictation and narration in 646 languages · Activity score 5.0/5 (Thriving) · Python · ★ 109945 - [google-gemini/gemini-cli](../google-gemini-gemini-cli.md): Terminal AI agent for Gemini Code Assist enterprise licenses · Activity score 5.0/5 (Thriving) · TypeScript · ★ 108820 - [vllm-project/vllm](../vllm-project-vllm.md): High-throughput inference and serving engine for LLMs with an OpenAI-compatible API · Activity score 5.0/5 (Thriving) · Python · ★ 96871 - [infiniflow/ragflow](../infiniflow-ragflow.md): RAG engine that turns PDFs and office documents into a citable context layer for LLMs · Activity score 5.0/5 (Thriving) · Go · ★ 94127 - [OpenHands/OpenHands](../openhands-openhands.md): Self-hosted control center for coding agents and scheduled automations · Activity score 5.0/5 (Thriving) · TypeScript · ★ 93424 - [modelcontextprotocol/servers](../modelcontextprotocol-servers.md): Reference MCP servers for files, Git, fetch, memory, time, and thinking · Activity score 5.0/5 (Thriving) · TypeScript · ★ 91973 - [cheaper-inference/OmniRoute](../cheaper-inference-omniroute.md): Self-hosted AI gateway that routes one endpoint across 375 providers · Activity score 5.0/5 (Thriving) · TypeScript · ★ 87164 - [diegosouzapw/OmniRoute](../diegosouzapw-omniroute.md): Never stop coding. · Activity score 5.0/5 (Thriving) · TypeScript · ★ 86687 - [heygen-com/hyperframes](../heygen-com-hyperframes.md): Render HTML, CSS and seekable animations into deterministic MP4 video · Activity score 5.0/5 (Thriving) · TypeScript · ★ 83925 - [BerriAI/litellm](../berriai-litellm.md): Self-hosted AI gateway and Python SDK for 100+ LLM providers · Activity score 5.0/5 (Thriving) · Python · ★ 63220 - [vectorize-io/hindsight](../vectorize-io-hindsight.md): Agent memory system built so agents learn from experience, not just recall · Activity score 5.0/5 (Thriving) · Python · ★ 60887 - [crewAIInc/crewAI](../crewaiinc-crewai.md): Python framework for role-based multi-agent crews and event-driven flows · Activity score 5.0/5 (Thriving) · Python · ★ 60863 - [pingdotgg/t3code](../pingdotgg-t3code.md): One interface for coding agents, with desktop, web and phone clients · Activity score 5.0/5 (Thriving) · TypeScript · ★ 59541 - [CherryHQ/cherry-studio](../cherryhq-cherry-studio.md): Desktop AI workspace for models, agents, knowledge bases and images · Activity score 5.0/5 (Thriving) · TypeScript · ★ 53489 - [HKUDS/nanobot](../hkuds-nanobot.md): Lightweight self-hosted personal AI agent in Python, in chat or browser · Activity score 5.0/5 (Thriving) · Python · ★ 50067 - [JustVugg/colibri](../justvugg-colibri.md): Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. · Activity score 5.0/5 (Thriving) · C · ★ 50060 - [AlexsJones/llmfit](../alexsjones-llmfit.md): Terminal tool that matches open LLMs to your CPU, RAM and GPU before you download them · Activity score 5.0/5 (Thriving) · Rust · ★ 38975 - [github/github-mcp-server](../github-github-mcp-server.md): GitHub's official MCP server for repositories, issues, PRs, and Actions · Activity score 5.0/5 (Thriving) · Go · ★ 34443 - [topoteretes/cognee](../topoteretes-cognee.md): AI memory platform that turns documents, code and conversations into a searchable knowledge graph · Activity score 5.0/5 (Thriving) · Python · ★ 33296 - [Kilo-Org/kilocode](../kilo-org-kilocode.md): Open-source coding agent for VS Code, JetBrains and the command line with 500+ models · Activity score 5.0/5 (Thriving) · TypeScript · ★ 27978 - [getpaseo/paseo](../getpaseo-paseo.md): Orchestrate multiple coding agents from desktop and mobile · Activity score 5.0/5 (Thriving) · TypeScript · ★ 23861 - [vercel-labs/agent-desktop](../vercel-labs-agent-desktop.md): Computer-use CLI that lets AI agents read and control desktop apps through the accessibility tree · Activity score 5.0/5 (Thriving) · Rust · ★ 23269 - [dyad-sh/dyad](../dyad-sh-dyad.md): Free, local, open-source AI app builder | v0 / lovable / Bolt alternative | 🌟 Star if you like it! · Activity score 5.0/5 (Thriving) · TypeScript · ★ 22269 - [livekit/agents](../livekit-agents.md): Python framework for realtime voice and multimodal AI agents on LiveKit WebRTC · Activity score 5.0/5 (Thriving) · Python · ★ 15215 - [NoeFabris/opencode-antigravity-auth](../noefabris-opencode-antigravity-auth.md): OpenCode plugin that authenticates Antigravity and Gemini CLI accounts over OAuth · Activity score 5.0/5 (Thriving) · TypeScript · ★ 11042 - [open-pencil/open-pencil](../open-pencil-open-pencil.md): Open-source design editor that opens Figma files and includes AI tools · Activity score 5.0/5 (Thriving) · TypeScript · ★ 9834 - [junhoyeo/tokscale](../junhoyeo-tokscale.md): A CLI tool for tracking token usage from OpenCode, Claude Code, 🦞OpenClaw, Pi, Codex, Gemini, Cursor, AmpCode, Factory Droid · Activity score 5.0/5 (Thriving) · Rust · ★ 5840 - [rivet-dev/agentos](../rivet-dev-agentos.md): WebAssembly sandboxes that give AI agents files, shell and networking · Activity score 5.0/5 (Thriving) · Rust · ★ 4894 - [neuron-core/neuron-ai](../neuron-core-neuron-ai.md): PHP framework for building and orchestrating AI agents · Activity score 5.0/5 (Thriving) · PHP · ★ 2174 - [RunanywhereAI/wally](../runanywhereai-wally.md): Get up and running with GLM-5.3-flash, DeepSeek, Gemma and other open source frontier models · Activity score 5.0/5 (Thriving) · Rust · ★ 1570 - [tintinweb/pi-subagents](../tintinweb-pi-subagents.md): Claude Code like Sub-agents for Pi — parallel execution, live widget, custom agent types, mid-run steering and more .. · Activity score 5.0/5 (Thriving) · TypeScript · ★ 1438 - [coollabsio/jean](../coollabsio-jean.md): Desktop workspace running coding agents side by side in separate git worktrees · Activity score 5.0/5 (Thriving) · TypeScript · ★ 1417 - [aicodeking/pandacode](../aicodeking-pandacode.md): Native desktop workspace that runs several coding agent CLIs side by side · Activity score 5.0/5 (Thriving) · TypeScript · ★ 1240 - [livekit/agents-js](../livekit-agents-js.md): Build realtime multimodal AI agents with Node.js. · Activity score 5.0/5 (Thriving) · TypeScript · ★ 963 - [agent-sh/computer-use-linux](../agent-sh-computer-use-linux.md): Gives AI agents structured control of Linux desktops over MCP, CLI and native tools · Activity score 5.0/5 (Thriving) · Rust · ★ 871 - [squirrelscan/squirrelscan](../squirrelscan-squirrelscan.md): CLI and MCP website audit tool that hands fixes to coding agents · Activity score 5.0/5 (Thriving) · TypeScript · ★ 312 - [future-agi/traceAI](../future-agi-traceai.md): Open Source AI Tracing Framework built on Opentelemetry for AI Applications and Frameworks · Activity score 5.0/5 (Thriving) · Python · ★ 266 - [octopusreview/octopus](../octopusreview-octopus.md): AI-powered PR review tool — multi-armed code analysis, no bug escapes 🐙 · Activity score 5.0/5 (Thriving) · TypeScript · ★ 115 - [ollama/ollama](../ollama-ollama.md): Local runtime that runs open-weight LLMs on macOS, Windows and Linux with one command · Activity score 4.9/5 (Thriving) · Go · ★ 186890 - [unclecode/crawl4ai](../unclecode-crawl4ai.md): Crawl4AI: Open-source LLM Friendly Web Crawler & Scraper. Don't be shy, join here: https://discord.gg/jP8KfhDhyN · Activity score 4.9/5 (Thriving) · Python · ★ 87254 - [run-llama/llama_index](../run-llama-llama-index.md): Framework for building RAG and agent applications over your own documents · Activity score 4.9/5 (Thriving) · Python · ★ 52894 - [ogulcancelik/herdr](../ogulcancelik-herdr.md): agent multiplexer that lives in your terminal. · Activity score 4.9/5 (Thriving) · Rust · ★ 50449 - [alibaba/page-agent](../alibaba-page-agent.md): In-page JavaScript agent that controls web interfaces through natural language · Activity score 4.9/5 (Thriving) · TypeScript · ★ 29681 - [coleam00/Archon](../coleam00-archon.md): YAML workflow engine that runs AI coding agents through repeatable, gated delivery steps · Activity score 4.9/5 (Thriving) · TypeScript · ★ 23912 - [google/ax](../google-ax.md): Google's open agentic orchestration runtime. · Activity score 4.9/5 (Thriving) · Go · ★ 14615 - [MinishLab/semble](../minishlab-semble.md): Code search engine for agents that returns snippets instead of whole files · Activity score 4.9/5 (Thriving) · Python · ★ 6417 - [yifanfeng97/Hyper-Extract](../yifanfeng97-hyper-extract.md): CLI that turns documents into knowledge graphs and hypergraphs with LLMs · Activity score 4.9/5 (Thriving) · Python · ★ 4238 - [cocoindex-io/cocoindex-code](../cocoindex-io-cocoindex-code.md): AST-based semantic code search CLI that saves context for coding agents · Activity score 4.9/5 (Thriving) · Python · ★ 2980 - [supercorp-ai/supergateway](../supercorp-ai-supergateway.md): Bridges MCP stdio servers to SSE, WebSocket and streamable HTTP · Activity score 4.9/5 (Thriving) · TypeScript · ★ 2960 - [arabold/docs-mcp-server](../arabold-docs-mcp-server.md): Docs MCP Server: Enhance Your AI Coding Assistant · Activity score 4.9/5 (Thriving) · TypeScript · ★ 1870 - [agent0ai/tuxee](../agent0ai-tuxee.md): Small Debian container with a web UI for files and persistent terminals · Activity score 4.9/5 (Thriving) · JavaScript · ★ 404 - [builtbyberry/laravel-swarm](../builtbyberry-laravel-swarm.md): Multi-agent swarm orchestration for Laravel, built on Laravel AI. · Activity score 4.9/5 (Thriving) · PHP · ★ 35 - [ggml-org/llama.cpp](../ggml-org-llama-cpp.md): C/C++ inference engine that runs quantized LLMs and vision models on CPU and GPU · Activity score 4.8/5 (Thriving) · C++ · ★ 134864 - [tesseract-ocr/tesseract](../tesseract-ocr-tesseract.md): OCR engine and command-line tool with LSTM models for many languages · Activity score 4.8/5 (Thriving) · C++ · ★ 77500 - [Niko1221/Strata](../niko1221-strata.md): Run very large MoE models locally with SSD offloading and a local API · Activity score 4.8/5 (Thriving) · C++ · ★ 43898 - [iOfficeAI/AionUi](../iofficeai-aionui.md): Free, local, open-source Cowork for Gemini CLI, Claude Code, Codex, Opencode, Qwen Code, Goose Cli, Auggie, and more | 🌟 Star · Activity score 4.8/5 (Thriving) · TypeScript · ★ 34229 - [experientiallabs/experiential](../experientiallabs-experiential.md): Open-source LLM gateway that routes hosted, BYOK and local models through one API · Activity score 4.8/5 (Thriving) · Python · ★ 16897 - [hardbeat920/monocode](../hardbeat920-monocode.md): Desktop GUI for running several coding agent sessions side by side · Activity score 4.8/5 (Thriving) · Rust · ★ 3955 - [prism-php/prism](../prism-php-prism.md): Unified Laravel interface for working with large language models · Activity score 4.8/5 (Thriving) · PHP · ★ 2430 - [openai/codex](../openai-codex.md): OpenAI's open-source coding agent that runs locally in the terminal · Activity score 4.7/5 (Thriving) · Rust · ★ 135182 - [craft-ai-agents/craft-agents-oss](../craft-ai-agents-craft-agents-oss.md): Desktop workspace where agents connect to APIs, MCP servers and sessions · Activity score 4.7/5 (Thriving) · TypeScript · ★ 7333 - [oneglanse/oneglanse](../oneglanse-oneglanse.md): Self-hosted tracker for how AI assistants mention and cite a brand · Activity score 4.7/5 (Thriving) · TypeScript · ★ 257 - [OpenMontageInc/OpenMontage](../openmontageinc-openmontage.md): Agentic video production pipelines driven from a coding assistant · Activity score 4.6/5 (Thriving) · Python · ★ 80280 - [phodal/routa](../phodal-routa.md): Workspace for coordinating several coding agents around shared specs and a board · Activity score 4.6/5 (Thriving) · TypeScript · ★ 1839 - [BeaconBay/ck](../beaconbay-ck.md): Local semantic and hybrid code search for developers and AI agents · Activity score 4.6/5 (Thriving) · Rust · ★ 1771 - [dappros/ethora](../dappros-ethora.md): Chat and messaging SDKs with AI bots for web, mobile and WordPress. · Activity score 4.6/5 (Thriving) · JavaScript · ★ 553 - [stew675/llama-cpp-rdna-boosts](../stew675-llama-cpp-rdna-boosts.md): Patch series that tunes llama.cpp inference for AMD RDNA GPUs · Activity score 4.6/5 (Thriving) · Shell · ★ 249 - [QwenLM/open-computer-use](../qwenlm-open-computer-use.md): Desktop and browser control runtimes that let AI agents operate a computer · Activity score 4.4/5 (Active) · Rust · ★ 319 - [mindsdb/mindshub](../mindsdb-mindshub.md): Agent workspace for knowledge work and software development with open-source harnesses and model choice · Activity score 4.3/5 (Active) · Makefile · ★ 39864 - [virattt/dexter](../virattt-dexter.md): Autonomous agent that plans and runs deep financial research with live market data · Activity score 4.2/5 (Active) · TypeScript · ★ 27832 - [Zagitta/gpu-unlock](../zagitta-gpu-unlock.md): Removes the tensor-core cap on GeForce cards to raise FP16, BF16 and TF32 throughput · Activity score 4.1/5 (Active) · C · ★ 19227 - [hexabot-ai/Hexabot](../hexabot-ai-hexabot.md): AI workflow automation platform combining agents, actions and chat channels · Activity score 4.0/5 (Active) · TypeScript · ★ 1161 - [hexastack/hexabot](../hexastack-hexabot.md): Hexabot is an open-source AI chatbot / agent builder. · Activity score 4.0/5 (Active) · TypeScript · ★ 1157 - [ImDarkTom/LlamaPen](../imdarktom-llamapen.md): No-install web GUI for Ollama and other local LLM servers, with local chat history · Activity score 4.0/5 (Active) · Vue · ★ 451 - [Assisted-Mindfulness/naive-bayes](../assisted-mindfulness-naive-bayes.md): Naive Bayes text classifier for PHP with configurable precision and tokenizer · Activity score 3.6/5 (Active) · PHP · ★ 45 - [MaestroError/LarAgent](../maestroerror-laragent.md): Power of AI Agents in your Laravel project · Activity score 3.5/5 (Active) · PHP · ★ 645 - [rishikanthc/Scriberr](../rishikanthc-scriberr.md): Self-hosted transcription for audio and video with speaker detection · Activity score 3.3/5 (Maintained) · Go · ★ 3146 - [memvid/memvid](../memvid-memvid.md): Single-file memory layer for AI agents with instant retrieval and long-term recall · Activity score 3.2/5 (Maintained) · Rust · ★ 17410 - [kelindar/search](../kelindar-search.md): Go library for embedded vector search and semantic embeddings using llama.cpp · Activity score 3.2/5 (Maintained) · Go · ★ 562 - [Alph4d0g/opencode-omniroute-auth](../alph4d0g-opencode-omniroute-auth.md): opencode authentication provider to authenticate against omniroute · Activity score 3.0/5 (Maintained) · TypeScript · ★ 151 - [Portkey-AI/gateway](../portkey-ai-gateway.md): Lightweight AI gateway with routing, guardrails, and 1600+ models · Activity score 2.9/5 (Maintained) · TypeScript · ★ 13378 - [AutoMaker-Org/automaker](../automaker-org-automaker.md): Learn more about Agentic Coding! · Activity score 2.9/5 (Maintained) · TypeScript · ★ 3223 - [pulp-vision/Hush](../pulp-vision-hush.md): 8 MB speech enhancement model that isolates a speaker in real time · Activity score 2.7/5 (Maintained) · Python · ★ 125 - [openai/whisper](../openai-whisper.md): General-purpose speech recognition model for transcription and translation · Activity score 2.6/5 (Maintained) · Python · ★ 112662 - [pushpak1300/ai-chat](../pushpak1300-ai-chat.md): Laravel starter kit for AI chat apps with streaming responses and multiple providers · Activity score 2.3/5 (Slowing down) · PHP · ★ 382 - [MorDavid/BruteForceAI](../mordavid-bruteforceai.md): LLM-assisted brute-force testing for login pages, built for authorised audits · Activity score 1.9/5 (Slowing down) · Python · ★ 1934 - [sivakumar-mahalingam/fastmrz](../sivakumar-mahalingam-fastmrz.md): Extracting the Machine Readable Zone (MRZ) from passport or any document images · Activity score 1.7/5 (Slowing down) · Python · ★ 173 - [raksbisht/sitechat](../raksbisht-sitechat.md): Open-source, self-hosted AI chatbot trained on your website and docs. RAG, embeddable widget, no vendor lock-in. · Activity score 1.7/5 (Slowing down) · Python · ★ 44 - [SciPhi-AI/R2R](../sciphi-ai-r2r.md): SoTA production-ready AI retrieval system. Agentic Retrieval-Augmented Generation (RAG) with a RESTful API. · Activity score 1.6/5 (Abandoned) · Python · ★ 8087 - [fictionco/fiction](../fictionco-fiction.md): Personal marketing platform for building sites, funnels and newsletter emails · Activity score 1.5/5 (Abandoned) · TypeScript · ★ 1471 - [insaaniManav/prompt-forge](../insaanimanav-prompt-forge.md): Workbench for generating, analysing and testing AI prompts · Activity score 1.5/5 (Abandoned) · Go · ★ 791 - [open-mmlab/mmdetection](../open-mmlab-mmdetection.md): OpenMMLab toolbox for object detection, instance segmentation and panoptic segmentation in PyTorch · Activity score 1.4/5 (Abandoned) · Python · ★ 33039 - [getumbrel/llama-gpt](../getumbrel-llama-gpt.md): Self-hosted, offline ChatGPT-style chatbot running Llama 2 models · Activity score 1.4/5 (Abandoned) · TypeScript · ★ 10923 - [gorgonia/gorgonia](../gorgonia-gorgonia.md): Machine learning library for Go with computation graphs and autodiff · Activity score 1.4/5 (Abandoned) · Go · ★ 5927 - [open-mmlab/mmocr](../open-mmlab-mmocr.md): OpenMMLab Text Detection, Recognition and Understanding Toolbox · Activity score 1.4/5 (Abandoned) · Python · ★ 4746 - [op7418/ai-claude-start](../op7418-ai-claude-start.md): Claude Code launcher that switches API profiles and keeps tokens in the OS keychain · Activity score 1.4/5 (Abandoned) · TypeScript · ★ 224 - [deepseek-ai/DeepSeek-OCR](../deepseek-ai-deepseek-ocr.md): Vision-language model for optical character recognition and document parsing via vLLM or Transformers · Activity score 1.3/5 (Abandoned) · Python · ★ 24157 - [coleam00/ottomator-agents](../coleam00-ottomator-agents.md): All the open source AI Agents hosted on the oTTomator Live Agent Studio platform! · Activity score 1.2/5 (Abandoned) · Python · ★ 5800 - [hieunc229/copilot-clone](../hieunc229-copilot-clone.md): VSCode extension for code suggestion. · Activity score 1.2/5 (Abandoned) · TypeScript · ★ 1782 - [manhcuong02/eKYC](../manhcuong02-ekyc.md): Python eKYC demo with ID card parsing, face matching and liveness checks · Activity score 1.0/5 (Abandoned) · Python · ★ 64