best-of-ai
Health Uyari
- License — License: MIT
- Description — Repository has a description
- Active repo — Last push 0 days ago
- Low visibility — Only 5 GitHub stars
Code Gecti
- Code scan — Scanned 5 files during light audit, no dangerous patterns found
Permissions Gecti
- Permissions — No dangerous permissions requested
Bu listing icin henuz AI raporu yok.
Open-source AI, ranked: apps, agents, models and the tools around them, each with strengths, weaknesses and a score from real GitHub signals. Re-checked twice a day.
Best of Open-Source AI
Open-source AI, ranked. 🏆
AI apps, agents and models you run, and the parts to build with. Paid never moves a rank.
Best of Open-Source AI is a ranked list of 844 open-source AI projects: chat apps and assistants, desktop apps, agent platforms and frameworks, workflow automation, RAG, memory, model serving, open models, training, image, video and voice, vector databases, gateways, observability, sandboxes, data tools, MCP servers and learning material. Each entry has strengths, weaknesses, specs and a score from public GitHub signals. Updated 2026-11-01 (UTC).
[!TIP]
🔎 Every project fact-checked on GitHub before it is listed · ✍️ Every entry has strengths and weaknesses · ▶️ Demo links so you can try before you deploy
[!NOTE]
🤝 Independent. We sell none of the tools listed here, and nobody can pay to move up. Ranks come from public GitHub data and are re-checked twice a day; hover a score to see what it is made of.
Looking for coding agents, starters and the stack you ship on? 👉 Best of Vibe Coding
🌐 The same ranking as a website: archestack.com, with a page per project and every score broken down.
🔥 Top 10 right now
844 projects ranked in 21 categories · data from 2026-11-01 · re-ranked twice a day
How to read it: rank 🥇🥈🥉 then #4, #5… (equal scores: more stars first) · score out of 100 sets the order: hover it to see what it is made of, or read how we rank · click a project name for its review · ↗ opens another site (Cmd/Ctrl-click for a new tab)
| Rank | Project | Score | Category | Stars |
|---|---|---|---|---|
| 🥇 | voicestudio Desktop app for voice cloning, dubbing, dictation and audiobooks 🐳 Docker · AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
97 | 🎙️ Voice | 97k |
| 🥈 | OmniRoute LLM gateway routing one OpenAI-compatible endpoint across 372 providers 🐳 Docker · MIT · 🌐 Site ↗ |
96 | 🔀 Gateways and observability | 83k |
| 🥉 | pytorch Python tensor library with GPU support and tape-based autograd for deep learning 🐳 Docker · custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 🔬 ML frameworks and libraries | 105k |
| #4 | ruview WiFi CSI sensing on ESP32 for presence, breathing, heart rate and pose 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 🧬 Open models | 102k |
| #5 | career-ops Job-search agent that scores listings, tailors CVs and drafts applications 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 🤖 Assistants | 79k |
| #6 | ppt-master Agent skill that turns documents into natively editable PowerPoint decks MIT · ▶️ Demo ↗ · 📖 Docs ↗ |
95 | 🧰 Agent frameworks and SDKs | 72k |
| #7 | ai-job-search Claude Code workflow that searches job boards and drafts tailored CVs and cover letters MIT |
95 | 🤖 Assistants | 69k |
| #8 | n8n Visual workflow automation with AI agent nodes and 1500+ integrations 🐳 Docker · custom license · 📖 Docs ↗ |
94 | 🔁 Workflow automation with AI | 209k |
| #9 | AutoGPT Role-based AI agents coordinated by one lead agent, plus a visual workflow builder 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 🧩 Agent platforms | 188k |
| #10 | transformers Python library defining open models for text, vision, audio and multimodal tasks 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 🔬 ML frameworks and libraries | 168k |
Ordered by score: adoption, freshness, how easy it is to run, and a few more signals as we measure them (how we rank).
📈 Rising this week
Most GitHub stars gained from October 25 to 31, 2026 (UTC), from GitHub's daily star counts · top 10, this month and an RSS feed: archestack.com/rising
| Project | Stars this week | Rank |
|---|---|---|
| voicestudio | +11,964 (+14%) | 🥇 in Voice |
| deeptutor | +10,623 (+25%) | 🥉 in RAG, knowledge and search |
| ifixai | +7,997 (+23%) | 🥈 in Security |
| tradingagents | +5,488 (+4.8%) | #5 in Agent frameworks and SDKs |
| strata | +4,992 (+15%) | #6 in Model serving |
⚡ Pick by vibe
- 💬 ChatGPT, but at home → Chat UIs
- 🖥️ An AI app on my laptop, offline → Desktop AI apps
- 🧠 Run models on my own GPU → Model serving
- 🤖 An assistant in my Telegram or WhatsApp → Assistants
- 📚 Chat with my own documents → RAG, knowledge and search
- 🧰 Build an agent in code → Agent frameworks and SDKs
- 🔁 Wire AI into the rest of my tools → Workflow automation with AI
- 🔀 One API key for every provider → Gateways and observability
- 🏋️ Fine-tune a model on my data → Training and fine-tuning
- 🕸️ Turn websites and PDFs into clean text → Data and scraping for AI
- 🔌 Give my assistant tools → MCP servers
- 📈 See what my LLM app is actually doing → Gateways and observability
- 🛡️ Let agents run code without wrecking my server → Sandboxes
- 📖 Learn to build with AI → Learning and cookbooks
🔓 Open-source alternatives to …
The top open-source picks by score in the matching categories. ✋ means the project's own README calls it an alternative to that product.
💻 Runs on what you have
From each project README: whether it needs a GPU, and whether it mentions Apple Silicon or arm64 or ships an arm64 build or container image. Apps and tools you run, not libraries.
| What you have | Projects | Top picks |
|---|---|---|
| No GPU needed | 189 | OmniRoute · career-ops · ai-job-search · n8n · servers · daily_stock_analysis |
| A GPU helps, not required | 176 | voicestudio · mineru · llama.cpp · openhuman · OpenClaw · cherry-studio |
| Needs a GPU | 46 | strata · tensorrt-llm · omlx · freetoken · KTransformers · web-llm |
| A Mac or another arm64 machine | 224 | voicestudio · OmniRoute · AutoGPT · mineru · hindsight · Langflow |
🗂️ All categories
🧑💻 Apps you run
- 🤖 Assistants · 37
- 💬 Chat UIs · 21
- 🖥️ Desktop AI apps · 24
- 🧩 Agent platforms · 52
- 🔁 Workflow automation with AI · 2
- 📚 RAG, knowledge and search · 42
- 🎙️ Voice · 67
- 🎨 Image and video · 42
- 🔐 Security · 14
🧱 Build with it
- 🧰 Agent frameworks and SDKs · 82
- 🗂️ Memory · 18
- 🧠 Model serving · 57
- 🧬 Open models · 52
- 🏋️ Training and fine-tuning · 41
- 🔬 ML frameworks and libraries · 126
- 🔀 Gateways and observability · 42
- 🧮 Vector databases · 15
- 🛡️ Sandboxes · 11
- 🕸️ Data and scraping for AI · 33
- 🔌 MCP servers · 1
📖 Learn
- 📖 Learning and cookbooks · 65
🧑💻 Apps you run
🤖 Assistants
Personal AI assistants you run yourself and talk to through chat apps, with memory and the ability to act. 37 ranked · 📝 all reviews · 🌐 on archestack.com
career-ops leads with 95, ahead of ai-job-search (95) and openhuman (93).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | career-ops Job-search agent that scores listings, tailors CVs and drafts applications 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 79k |
| 🥈 | ai-job-search Claude Code workflow that searches job boards and drafts tailored CVs and cover letters MIT |
95 | 69k |
| 🥉 | openhuman Rust agent harness with desktop app, built-in memory and embeddable library 🐳 Docker · GPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
93 | 42k |
| #4 | OpenClaw Personal AI assistant reachable through Discord, Slack, Telegram, WhatsApp and more 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
92 | 394k |
| #5 | airi AI VTuber companion with VRM and Live2D avatars, voice, and game-playing MIT · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
91 | 51k |
| #6 | Hermes Agent Self-hosted agent with skill learning, memory and Telegram/Discord/Slack gateway 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
87 | 259k |
| #7 | nanobot Self-hosted personal AI agent reachable from browser, terminal and chat apps 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 50k |
| #8 | AstrBot Agent chatbot platform connecting LLMs to QQ, Telegram, Slack and other chat apps 🐳 Docker · AGPL-3.0 · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 45k |
| #9 | cowagent Agent assistant you chat with via web console, Telegram, Slack and more 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 48k |
| #10 | ZeroClaw Single Rust binary agent runtime with 30+ channels and hardware access 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 33k |
Top 10 of 37. See all 37, ranked, with reviews →
💡 How to choose- Check which channels it supports today (WhatsApp, Telegram, Slack, Discord, email) and whether each needs a paid API.
- Look at how memory is stored and whether you can inspect or wipe it.
- Actions need credentials; prefer assistants that scope them per tool.
💬 Chat UIs
Web front-ends for local or API models, usually with user accounts, chat history and file upload. 21 ranked · 📝 all reviews · 🌐 on archestack.com
Open WebUI leads with 89, ahead of LobeHub (86) and Onyx (85).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | Open WebUI Self-hosted web UI for Ollama and OpenAI-compatible models with users and RAG 🐳 Docker · custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 157k |
| 🥈 | LobeHub Agent workspace with builder, groups, scheduling and 10,000+ MCP skills 🐳 Docker · custom license |
86 | 84k |
| 🥉 | Onyx Team knowledge search and chat that indexes 50+ connected apps with permissions 🐳 Docker · custom license · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 33k |
| #4 | AnythingLLM Document chat app with agents, workspaces and many LLM and vector DB backends 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 68k |
| #5 | LibreChat Self-hosted multi-provider chat UI with agents, MCP tools and multi-user auth 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 47k |
| #6 | Hermes WebUI Browser front-end for Hermes Agent with sessions, workspace files and profiles 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
76 | 19k |
| #7 | odysseus Self-hosted workspace for chat, agents, email, documents and calendar 🐳 Docker · AGPL-3.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
71 | 95k |
| #8 | SillyTavern Local chat front end for role-play across many LLM backends 🐳 Docker · AGPL-3.0 · 📖 Docs ↗ |
70 | 35k |
| #9 | ekko-studio Desktop and web workspace for chatting with Hermes, Ekko and coding agents 🐳 Docker · custom license · 📖 Docs ↗ · 🌐 Site ↗ |
70 | 12k |
| #10 | big-AGI Multi-model chat workspace with Beam side-by-side model comparison 🐳 Docker · MIT · 🌐 Site ↗ |
69 | 7.2k |
Top 10 of 21. See all 21, ranked, with reviews →
💡 How to choose- Confirm it talks to your backend (Ollama, OpenAI-compatible, Anthropic) without a plugin.
- Multi-user auth, RBAC and SSO are where free and paid editions differ most.
- Check what the default compose pulls in (database, vector store) before sizing the host.
🖥️ Desktop AI apps
Apps you install on your own computer to chat with local or API models, often with document chat and agents built in. 24 ranked · 📝 all reviews · 🌐 on archestack.com
cherry-studio leads with 92, ahead of dsh-desktop (88) and screenpipe (88).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | cherry-studio Desktop client for chatting with many LLM providers, with agents and knowledge bases AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
92 | 53k |
| 🥈 | dsh-desktop Native Windows and macOS desktop client for DeepSeek Harness, built on plugins MIT · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 34k |
| 🥉 | screenpipe Records screen and audio locally, makes it searchable for you and your agents custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 22k |
| #4 | prompt-optimizer Prompt rewriting and testing tool for web, desktop, Chrome and MCP 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 40k |
| #5 | deep-live-cam Real-time webcam face swap and video deepfake from a single source image AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 98k |
| #6 | aionui Desktop and WebUI app that runs a built-in agent plus external CLI agents 🐳 Docker · Apache-2.0 · 🌐 Site ↗ |
84 | 34k |
| #7 | jan Desktop app that runs local models or cloud APIs, with a local server custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 45k |
| #8 | hermes-desktop Desktop GUI for installing and chatting with Hermes Agent MIT · GPU optional · 🌐 Site ↗ |
82 | 15k |
| #9 | chatbox Desktop and mobile chat client for OpenAI, Claude, Gemini and Ollama models GPL-3.0 · 🌐 Site ↗ |
79 | 42k |
| #10 | gallery Mobile app for running open LLMs offline on Android and iOS Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
75 | 25k |
Top 10 of 24. See all 24, ranked, with reviews →
💡 How to choose- Check which model backends work out of the box: a built-in runtime, Ollama, or API keys.
- See whether chats and files stay on disk or sync to a cloud account.
- Note the platforms: several are macOS-first.
🧩 Agent platforms
Visual or code-first builders for agents and workflows, with orchestration, tools and deployment. 52 ranked · 📝 all reviews · 🌐 on archestack.com
AutoGPT leads with 94, ahead of Langflow (93) and daily_stock_analysis (93).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | AutoGPT Role-based AI agents coordinated by one lead agent, plus a visual workflow builder 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 188k |
| 🥈 | Langflow Visual builder for AI agents and workflows, exposed as API or MCP server 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
93 | 156k |
| 🥉 | daily_stock_analysis LLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps 🐳 Docker · MIT · 📖 Docs ↗ |
93 | 67k |
| #4 | Dify Visual builder for LLM workflows, RAG pipelines and agents, with APIs 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
92 | 160k |
| #5 | open-design Design workspace that drives your coding agent to build prototypes, decks and video 🐳 Docker · Apache-2.0 · 🌐 Site ↗ |
92 | 105k |
| #6 | resume-matcher Tailors resumes and cover letters to job descriptions using local or hosted LLMs 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 29k |
| #7 | deer-flow Agent harness running sub-agents, memory and sandboxes with pluggable skills 🐳 Docker · MIT · 🌐 Site ↗ |
88 | 85k |
| #8 | Multica Issue board where coding agent CLIs are assigned work like teammates 🐳 Docker · custom license · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 54k |
| #9 | suna Git-repo-based agent platform running each session in its own cloud sandbox custom license · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 20k |
| #10 | FastGPT Knowledge-base Q&A and visual workflow platform for LLM apps 🐳 Docker · custom license · 📖 Docs ↗ · 🌐 Site ↗ |
87 | 30k |
Top 10 of 52. See all 52, ranked, with reviews →
💡 How to choose- Decide between a visual builder (faster to start) and code-first (easier to test and version).
- Check how workflows are exported; vendor-specific JSON makes migration costly.
- Look at the license for the server part; several are AGPL or source-available.
🔁 Workflow automation with AI
Workflow and integration platforms with AI steps or agent nodes, for wiring models into the rest of your tools. 2 ranked · 📝 all reviews · 🌐 on archestack.com
n8n leads with 94, ahead of Activepieces (75).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | n8n Visual workflow automation with AI agent nodes and 1500+ integrations 🐳 Docker · custom license · 📖 Docs ↗ |
94 | 209k |
| 🥈 | Activepieces Self-hosted workflow automation with TypeScript integrations, alternative to Zapier 🐳 Docker · custom license · 📖 Docs ↗ · 🌐 Site ↗ |
75 | 25k |
- Count the integrations you actually need; the long tail is where platforms differ.
- Check whether AI steps can use your own model endpoint or only hosted providers.
- Read the license: several are fair-code or source-available, not OSI open source.
📚 RAG, knowledge and search
Document Q&A, knowledge bases and private search over your own files and the web, with queries kept on your host. 42 ranked · 📝 all reviews · 🌐 on archestack.com
RAGFlow leads with 90, ahead of LightRAG (89) and deeptutor (88).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | RAGFlow RAG engine with deep document parsing, template chunking and agentic retrieval 🐳 Docker · Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
90 | 93k |
| 🥈 | LightRAG Graph-based RAG framework with a REST API server and web UI 🐳 Docker · MIT · 📖 Docs ↗ |
89 | 40k |
| 🥉 | deeptutor Self-hosted AI tutor with knowledge bases, guided learning, books and quizzes 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 53k |
| #4 | MaxKB Enterprise knowledge-base agent platform with RAG, workflows and MCP tools 🐳 Docker · GPL-3.0 |
86 | 23k |
| #5 | quivr Go engine that ingests content streams for hybrid search and keyword alerts 🐳 Docker · MIT · 📖 Docs ↗ |
84 | 40k |
| #6 | Open Notebook Self-hosted NotebookLM alternative with podcasts and 20+ model providers 🐳 Docker · MIT · 🌐 Site ↗ |
83 | 41k |
| #7 | SurfSense Desktop NotebookLM-style agent that researches and edits local documents Apache-2.0 · GPU optional · 🌐 Site ↗ |
83 | 17k |
| #8 | pageindex Vectorless RAG that builds a tree index of a document and searches it with an LLM MIT · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 40k |
| #9 | GPT Researcher Research agent that writes cited reports from web and local documents 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
81 | 30k |
| #10 | WeKnora Knowledge base framework with RAG Q&A, tool-using agents and auto-generated wikis 🐳 Docker · custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
80 | 36k |
Top 10 of 42. See all 42, ranked, with reviews →
💡 How to choose- Match the ingestion formats you need (PDF with tables, Office, web, Confluence) before anything else.
- Check which embedding models and vector stores are supported and whether they can run offline.
- Look for citations in answers; without them, RAG output is hard to trust.
🎙️ Voice
Speech-to-text, text-to-speech, voice agents and meeting tools that run locally. 67 ranked · 📝 all reviews · 🌐 on archestack.com
voicestudio leads with 97, ahead of whisper.cpp (89) and handy (89).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | voicestudio Desktop app for voice cloning, dubbing, dictation and audiobooks 🐳 Docker · AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
97 | 97k |
| 🥈 | whisper.cpp C/C++ port of OpenAI Whisper for local speech-to-text transcription 🐳 Docker · MIT · GPU optional |
89 | 55k |
| 🥉 | handy Offline push-to-talk dictation app that pastes transcripts into any text field MIT · GPU optional · 🌐 Site ↗ |
89 | 36k |
| #4 | xiaozhi-esp32 ESP32 firmware for a voice chatbot with MCP device and cloud tool control 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 32k |
| #5 | buzz Offline desktop app for transcribing and translating audio with Whisper MIT · GPU optional · 📖 Docs ↗ |
81 | 22k |
| #6 | agents Python framework for realtime voice and multimodal agents on LiveKit WebRTC Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
79 | 15k |
| #7 | ebook2audiobook Converts ebooks to chaptered audiobooks with local TTS and voice cloning 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ |
78 | 20k |
| #8 | pyvideotrans Video translation and AI dubbing pipeline with ASR, subtitle translation and TTS 🐳 Docker · GPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
78 | 20k |
| #9 | sherpa-onnx Offline speech recognition, TTS, diarization and VAD runtime built on ONNX Runtime Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
78 | 16k |
| #10 | meetily Desktop app that records, transcribes and summarizes meetings locally MIT · GPU optional · ▶️ Demo ↗ · 🌐 Site ↗ |
77 | 32k |
Top 10 of 67. See all 67, ranked, with reviews →
💡 How to choose- Latency decides usability for voice agents; check the end-to-end numbers the project publishes.
- Most quality models need a GPU; CPU-only setups are slower and limited to smaller models.
- Check language coverage and whether models are downloaded at build or at first run.
🎨 Image and video
Generation UIs and pipelines for images and video, usually around diffusion models. 42 ranked · 📝 all reviews · 🌐 on archestack.com
frigate leads with 91, ahead of MoneyPrinterTurbo (87) and ComfyUI (86).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | frigate Local NVR with realtime object detection for IP cameras 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ |
91 | 37k |
| 🥈 | MoneyPrinterTurbo Generates short videos from a topic, with script, footage, voiceover and subtitles 🐳 Docker · MIT · GPU optional |
87 | 137k |
| 🥉 | ComfyUI Node-graph editor for running diffusion image, video, audio and 3D workflows GPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 144k |
| #4 | InvokeAI Web UI for diffusion image generation with inpainting canvas and node workflows 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 29k |
| #5 | toonflow-app Infinite-canvas tool for AI short drama and comic video production 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 20k |
| #6 | banana-slides AI slide deck generator that exports PPTX, PDF and narrated video 🐳 Docker · AGPL-3.0 · ▶️ Demo ↗ · 📖 Docs ↗ |
77 | 16k |
| #7 | open-generative-ai Image, video and lip-sync studio app that calls hosted models through MuAPI 🐳 Docker · MIT · GPU optional · ▶️ Demo ↗ · 🌐 Site ↗ |
72 | 32k |
| #8 | infinite-canvas Browser-based infinite canvas for AI image and video generation 🐳 Docker · MIT · 📖 Docs ↗ |
70 | 9.5k |
| #9 | stable-diffusion.cpp C/C++ inference for Stable Diffusion, Flux, Qwen-Image and Wan video models 🐳 Docker · MIT · GPU optional |
69 | 7.9k |
| #10 | ltx-2 Text and image to video model with synchronized audio, plus pipelines and trainer custom license · GPU required · ▶️ Demo ↗ · 🌐 Site ↗ |
68 | 10k |
Top 10 of 42. See all 42, ranked, with reviews →
💡 How to choose- A GPU with enough VRAM is the hard requirement; check the minimum the project states.
- Node-based UIs are flexible but have a learning curve; form-based UIs are quicker to use.
- Check the model licensing separately from the app licensing.
🔐 Security
AI agents and tools for penetration testing, red-teaming and finding vulnerabilities in your own apps and models. 14 ranked · 📝 all reviews · 🌐 on archestack.com
shannon leads with 91, ahead of ifixai (86) and Strix (84).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | shannon Autonomous pentester that exploits web apps and APIs using source code analysis 🐳 Docker · AGPL-3.0 · 📖 Docs ↗ · 🌐 Site ↗ |
91 | 51k |
| 🥈 | ifixai Python CLI that audits AI agents for policy, access and reliability failures Apache-2.0 · 🌐 Site ↗ |
86 | 42k |
| 🥉 | Strix Autonomous AI pentesting agents that validate vulnerabilities with working exploits 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 73k |
| #4 | pentagi Autonomous multi-agent penetration testing system with sandboxed security tools 🐳 Docker · MIT · GPU optional · 🌐 Site ↗ |
78 | 33k |
| #5 | ai-infra-guard Red-teaming platform for AI infra CVEs, MCP servers, agent skills and jailbreaks 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
71 | 7.3k |
| #6 | garak Command-line scanner that probes LLMs for jailbreaks, leakage and injection Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
68 | 9.8k |
| #7 | guardrails Programmable input, output and dialog guardrails for LLM applications 🐳 Docker · custom license · 📖 Docs ↗ |
68 | 7.4k |
| #8 | cyberstrikeai AI agent platform for authorized penetration testing Apache-2.0 · 📖 Docs ↗ |
66 | 8.3k |
| #9 | decepticon Autonomous red-team agents that run multi-stage attack chains from an approved plan 🐳 Docker · Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
66 | 6.5k |
| #10 | pentestgpt Autonomous pentest agent that drives Claude Code or Codex against a target 🐳 Docker · MIT · 🌐 Site ↗ |
53 | 16k |
Top 10 of 14. See all 14, ranked, with reviews →
💡 How to choose- Run them only against targets you own or are authorized to test; check the tool's scope controls.
- Check which model backends are supported and whether findings stay on your host.
- Look at how exploitation is sandboxed (containers, VMs) before pointing it at production.
🧱 Build with it
🧰 Agent frameworks and SDKs
Libraries for building agents and LLM apps in code: tool calling, planning, multi-agent setups and browser control. 82 ranked · 📝 all reviews · 🌐 on archestack.com
ppt-master leads with 95, ahead of openai-agents-python (90) and langchain (89).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | ppt-master Agent skill that turns documents into natively editable PowerPoint decks MIT · ▶️ Demo ↗ · 📖 Docs ↗ |
95 | 72k |
| 🥈 | openai-agents-python Python SDK for multi-agent workflows with handoffs, guardrails and tracing MIT · 📖 Docs ↗ |
90 | 30k |
| 🥉 | langchain Python framework for building LLM agents and apps from interoperable components MIT · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 149k |
| #4 | browser-use Python library that lets an LLM agent control a web browser 🐳 Docker · MIT · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 120k |
| #5 | tradingagents Multi-agent LLM framework that simulates a trading firm's analysis process 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ |
89 | 119k |
| #6 | copilotkit SDK for adding agent chat, generative UI and shared state to apps MIT · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 38k |
| #7 | mastra TypeScript framework for building AI agents, workflows and MCP servers custom license · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 29k |
| #8 | agent-browser Native Rust CLI that lets AI agents drive Chrome via accessibility snapshots 🐳 Docker · Apache-2.0 |
86 | 45k |
| #9 | agno Python SDK and runtime for building and serving agents behind a REST API Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 43k |
| #10 | crewai Python framework for multi-agent workflows with role-based Crews and event-driven Flows MIT · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 60k |
Top 10 of 82. See all 82, ranked, with reviews →
💡 How to choose- Pick by language first (Python or TypeScript), then by how much the framework hides.
- Check how it handles tool errors and retries; that is where agents break in production.
- Look for tracing hooks so you can see each step later.
🗂️ Memory
Long-term memory engines that store and retrieve facts for agents across sessions. 18 ranked · 📝 all reviews · 🌐 on archestack.com
hindsight leads with 94, ahead of MemPalace (90) and Mem0 (89).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | hindsight Agent memory server with retain, recall and reflect operations 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 57k |
| 🥈 | MemPalace Local verbatim memory store for AI agents with semantic search and MCP 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
90 | 60k |
| 🥉 | Mem0 Long-term memory layer that stores and retrieves user facts for LLM agents Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 68k |
| #4 | gbrain Long-term memory store for coding agents, with sources and corrections 🐳 Docker · MIT · 🌐 Site ↗ |
87 | 31k |
| #5 | Supermemory Memory and context engine for AI agents, with user profiles and hybrid search MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 31k |
| #6 | Cognee Memory engine that turns documents and code into a knowledge graph 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 33k |
| #7 | Graphiti Python framework for temporal knowledge graphs that track how facts change 🐳 Docker · Apache-2.0 · 🌐 Site ↗ |
84 | 32k |
| #8 | OpenViking Filesystem-style context database for agent memory, resources and skills 🐳 Docker · AGPL-3.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
81 | 41k |
| #9 | agentmemory Persistent cross-session memory for coding agents via hooks, MCP and REST 🐳 Docker · Apache-2.0 |
77 | 30k |
| #10 | everos Local-first agent memory runtime that stores memories as Markdown files Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
74 | 14k |
Top 10 of 18. See all 18, ranked, with reviews →
💡 How to choose- Check what gets stored (raw messages, extracted facts, graphs) and how it is retrieved.
- Multi-tenant isolation matters if several users or agents share one store.
- Look at the backing store (Postgres, a vector DB, a graph DB) and whether you already run it.
🧠 Model serving
Inference engines and model servers that expose local models over an API. 57 ranked · 📝 all reviews · 🌐 on archestack.com
llama.cpp leads with 93, ahead of Ollama (91) and LocalAI (90).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | llama.cpp C/C++ inference engine serving GGUF models over an OpenAI-compatible API 🐳 Docker · MIT · GPU optional · 🌐 Site ↗ |
93 | 133k |
| 🥈 | Ollama Local model runner with CLI and REST API for open-weight LLMs 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
91 | 186k |
| 🥉 | LocalAI Self-hosted inference server with OpenAI-compatible API and pluggable backends 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
90 | 50k |
| #4 | vLLM Inference engine for Hugging Face LLMs with PagedAttention and continuous batching 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 96k |
| #5 | colibri C inference engine that streams MoE experts from disk to run huge models 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 45k |
| #6 | strata Runs a 125B-parameter MoE model on one consumer GPU plus system RAM 🐳 Docker · MIT · GPU required |
85 | 37k |
| #7 | modular Mojo language and MAX inference framework with an OpenAI-compatible server custom license · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 30k |
| #8 | SGLang Inference engine for LLM, vision-language and diffusion models on many hardware types 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 38k |
| #9 | tensorrt-llm NVIDIA inference engine for LLMs and diffusion models with custom CUDA kernels 🐳 Docker · custom license · GPU required · 📖 Docs ↗ |
80 | 15k |
| #10 | airllm Python library that runs huge LLMs on small GPUs by streaming layers Apache-2.0 · GPU optional · ▶️ Demo ↗ |
79 | 41k |
Top 10 of 57. See all 57, ranked, with reviews →
💡 How to choose- Pick by hardware first (CPU, NVIDIA, AMD, Apple Silicon) and by model format (GGUF, safetensors).
- Throughput engines (continuous batching, paged attention) need GPUs; single-user servers do not.
- An OpenAI-compatible API keeps the rest of your stack portable.
🧬 Open models
Repositories of open-weight models with the code to run them: LLMs, vision, speech and embedding models. 52 ranked · 📝 all reviews · 🌐 on archestack.com
ruview leads with 95, ahead of llmfit (87) and laya (86).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | ruview WiFi CSI sensing on ESP32 for presence, breathing, heart rate and pose 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 102k |
| 🥈 | llmfit Rust tool that checks which LLMs fit your RAM, CPU and GPU 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
87 | 39k |
| 🥉 | laya Single-pass multilingual classifier for typed choice, score and yes/no decisions 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
86 | 35k |
| #4 | timesfm Pretrained time-series foundation model for zero-shot forecasting from Google Research Apache-2.0 · GPU optional · 🌐 Site ↗ |
77 | 35k |
| #5 | yolov5 PyTorch YOLOv5 object detection, segmentation and classification models with training code AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
75 | 58k |
| #6 | speech PyTorch toolkit for training and deploying ASR, TTS and speech LLM models 🐳 Docker · Apache-2.0 · GPU required · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
75 | 19k |
| #7 | rf-detr Real-time transformer models for object detection, segmentation and keypoints Apache-2.0 · GPU optional · ▶️ Demo ↗ · 🌐 Site ↗ |
74 | 9.9k |
| #8 | yue Lyrics-to-song model that writes an editable melody and chord score first Apache-2.0 · GPU required · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
71 | 13k |
| #9 | espnet PyTorch toolkit for speech recognition, synthesis, translation and enhancement 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
68 | 10k |
| #10 | needle 8-29 MB on-device model for tool calling, extraction and embeddings Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
67 | 14k |
Top 10 of 52. See all 52, ranked, with reviews →
💡 How to choose- Read the license: open weights are not always open for commercial use.
- Check the hardware the model needs at the precision you plan to run.
- See whether the repo ships inference code or only weights.
🏋️ Training and fine-tuning
Tools to fine-tune, train or distill models on your own data and GPUs, from LoRA scripts to full training stacks. 41 ranked · 📝 all reviews · 🌐 on archestack.com
unsloth leads with 88, ahead of peft (86) and deepspeed (85).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | unsloth Desktop and web app to run and fine-tune local models 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
88 | 79k |
| 🥈 | peft Python library for LoRA and other parameter-efficient fine-tuning of Transformers models 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 22k |
| 🥉 | deepspeed PyTorch library for distributed training of very large models with ZeRO 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 43k |
| #4 | verl Reinforcement learning training library for LLM post-training with PPO and GRPO 🐳 Docker · Apache-2.0 · GPU required · 📖 Docs ↗ |
84 | 24k |
| #5 | trl Post-training library with SFT, DPO, GRPO and KTO trainers 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ |
83 | 20k |
| #6 | megatron-lm Distributed training library for large transformer models on NVIDIA GPUs 🐳 Docker · custom license · GPU required · 📖 Docs ↗ |
80 | 18k |
| #7 | axolotl YAML-driven fine-tuning framework for LLMs with LoRA, QLoRA, RL and multi-GPU 🐳 Docker · Apache-2.0 · GPU required · ▶️ Demo ↗ · 📖 Docs ↗ |
79 | 13k |
| #8 | ms-swift Fine-tuning, RL, quantization and serving toolkit for 600+ text and 400+ multimodal LLMs Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
78 | 16k |
| #9 | heretic Removes refusal behavior from transformer language models via automated abliteration AGPL-3.0 · GPU optional |
77 | 37k |
| #10 | llamafactory Fine-tuning toolkit for 100+ LLMs with a CLI and Gradio web UI 🐳 Docker · Apache-2.0 · GPU required · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
75 | 76k |
Top 10 of 41. See all 41, ranked, with reviews →
💡 How to choose- Check the GPU memory it needs for your model size; LoRA and QLoRA change that a lot.
- See which model families and formats it supports in and out.
- Prefer tools with ready configs for your model over generic trainers.
🔬 ML frameworks and libraries
The libraries models are built and run with: tensor frameworks, model zoos and the tooling around them. 126 ranked · 📝 all reviews · 🌐 on archestack.com
pytorch leads with 95, ahead of transformers (94) and ultralytics (93).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | pytorch Python tensor library with GPU support and tape-based autograd for deep learning 🐳 Docker · custom license · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
95 | 105k |
| 🥈 | transformers Python library defining open models for text, vision, audio and multimodal tasks 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 168k |
| 🥉 | ultralytics Python library and CLI for training and running YOLO vision models 🐳 Docker · AGPL-3.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
93 | 63k |
| #4 | pytorch-image-models PyTorch library of image models, pretrained weights and reference training scripts Apache-2.0 · GPU optional · 📖 Docs ↗ |
93 | 37k |
| #5 | supervision Python toolkit for loading, annotating and counting computer vision detections MIT · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
91 | 51k |
| #6 | diffusers PyTorch library of diffusion pipelines and schedulers for image generation 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ |
85 | 35k |
| #7 | ray Distributed Python runtime with libraries for data, training, tuning, RL and serving 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 44k |
| #8 | mediapipe On-device ML task APIs and a C++ pipeline framework for mobile, web and desktop 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 37k |
| #9 | genesis-world Python multi-physics simulator for robotics and physical AI research Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
84 | 30k |
| #10 | mlx NumPy-style array framework for machine learning on Apple silicon MIT · GPU optional · 📖 Docs ↗ |
84 | 29k |
Top 10 of 126. See all 126, ranked, with reviews →
💡 How to choose- Pick the framework your target models and papers already use.
- Check hardware support (CUDA, ROCm, Apple Silicon, TPU) for your machines.
- Look at release cadence; a framework that stalls strands your code.
🔀 Gateways and observability
LLM gateways and proxies for routing, caching and cost control, and tracing, evaluation and prompt management for LLM apps. 42 ranked · 📝 all reviews · 🌐 on archestack.com
OmniRoute leads with 96, ahead of new-api (90) and MLflow (89).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | OmniRoute LLM gateway routing one OpenAI-compatible endpoint across 372 providers 🐳 Docker · MIT · 🌐 Site ↗ |
96 | 83k |
| 🥈 | new-api Gateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs 🐳 Docker · AGPL-3.0 · 📖 Docs ↗ |
90 | 51k |
| 🥉 | MLflow Tracing, evals, prompt registry and AI gateway plus classic ML tracking 🐳 Docker · Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 29k |
| #4 | LiteLLM OpenAI-format gateway and Python SDK for calling 100+ LLM providers 🐳 Docker · custom license · 📖 Docs ↗ · 🌐 Site ↗ |
87 | 62k |
| #5 | freellmapi OpenAI-compatible gateway pooling free-tier LLM providers with automatic failover 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
86 | 45k |
| #6 | Langfuse Tracing, prompt management and evals for LLM apps on ClickHouse 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 37k |
| #7 | Opik Trace, evaluate and monitor LLM apps and agents, Apache-2.0 end to end 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 23k |
| #8 | promptfoo CLI for evaluating and red-teaming prompts, agents and RAG 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 27k |
| #9 | experiential OpenAI-compatible LLM gateway with per-user budgets and traffic-trained routers Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
82 | 14k |
| #10 | llm-gateway OpenAI-compatible gateway routing requests across providers, subscriptions, local models 🐳 Docker · MIT · 📖 Docs ↗ · 🌐 Site ↗ |
79 | 7.6k |
Top 10 of 42. See all 42, ranked, with reviews →
💡 How to choose- List the providers you call today; check the gateway's native support rather than generic passthrough.
- Key management, per-team budgets and audit logs separate gateways from simple proxies.
- Check latency overhead and whether streaming is passed through unchanged.
🧮 Vector databases
Vector stores and hybrid search engines for embeddings. 15 ranked · 📝 all reviews · 🌐 on archestack.com
Milvus leads with 89, ahead of Meilisearch (85) and Qdrant (83).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | Milvus Distributed vector database with dense, sparse and hybrid search at scale 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
89 | 47k |
| 🥈 | Meilisearch Rust search engine with typo tolerance, filtering and hybrid semantic search 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 60k |
| 🥉 | Qdrant Rust vector database with payload filtering, REST and gRPC 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
83 | 35k |
| #4 | faiss C++ library for similarity search and clustering of dense vectors MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
80 | 41k |
| #5 | Weaviate Go vector database with built-in vectorizers, hybrid search and RAG 🐳 Docker · custom license · ▶️ Demo ↗ · 📖 Docs ↗ |
72 | 17k |
| #6 | lancedb Embedded vector database built on the Lance columnar format 🐳 Docker · Apache-2.0 · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
72 | 12k |
| #7 | Chroma Embedding database with a four-function API for Python and JavaScript 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
70 | 30k |
| #8 | zvec In-process vector database library with hybrid and full-text search Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
70 | 16k |
| #9 | leann Vector index that recomputes embeddings on demand to cut storage 🐳 Docker · MIT · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
69 | 14k |
| #10 | pgvector PostgreSQL extension for vector similarity search with HNSW and IVFFlat 🐳 Docker · custom license |
68 | 24k |
Top 10 of 15. See all 15, ranked, with reviews →
💡 How to choose- If you already run Postgres, check pgvector-based options before adding a new database.
- Hybrid search (keyword + vector) and filtering are where engines differ most.
- Check memory use per million vectors; it decides the host size.
🛡️ Sandboxes
Isolated runtimes where agents execute code, browse or use tools safely. 11 ranked · 📝 all reviews · 🌐 on archestack.com
Lightpanda leads with 92, ahead of OpenShell (86) and Obscura (85).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | Lightpanda Headless browser written in Zig for agents and scraping, with CDP and MCP 🐳 Docker · AGPL-3.0 · 📖 Docs ↗ · 🌐 Site ↗ |
92 | 40k |
| 🥈 | OpenShell Policy-enforced sandbox runtime for autonomous AI agents 🐳 Docker · Apache-2.0 · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ |
86 | 21k |
| 🥉 | Obscura Rust headless browser with V8 and CDP for scraping and agents 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
85 | 30k |
| #4 | NemoClaw Reference stack for running OpenClaw, Hermes and Deep Agents in OpenShell sandboxes 🐳 Docker · Apache-2.0 · 📖 Docs ↗ |
83 | 23k |
| #5 | opensandbox Sandbox API for running AI agent code, commands and browsers in isolation 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
81 | 16k |
| #6 | cubesandbox MicroVM sandbox service for AI agents with an E2B-compatible API 🐳 Docker · custom license |
74 | 13k |
| #7 | monty Embedded Python interpreter in Rust for running LLM-generated code MIT · 📖 Docs ↗ · 🌐 Site ↗ |
71 | 8.8k |
| #8 | microsandbox Embeddable local microVM sandboxes for running untrusted code and AI agents 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
67 | 8.9k |
| #9 | Steel Browser Browser API that manages Chrome sessions for Puppeteer, Playwright and Selenium 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
57 | 7.9k |
| #10 | Open Terminal REST API shell and file server for AI agents to run commands and code 🐳 Docker · MIT · 📖 Docs ↗ |
53 | 3.5k |
Top 10 of 11. See all 11, ranked, with reviews →
💡 How to choose- Isolation level (container, microVM, gVisor) sets the risk you accept when agents run code.
- Check startup time per sandbox; slow starts limit agent loops.
- Look at what is persisted between runs and how secrets are injected.
🕸️ Data and scraping for AI
Tools that turn web pages, PDFs and documents into clean text or structured data that models can use. 33 ranked · 📝 all reviews · 🌐 on archestack.com
mineru leads with 94, ahead of docling (91) and opendataloader-pdf (87).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | mineru Document parser that converts PDFs, images and Office files to Markdown and JSON 🐳 Docker · custom license · GPU optional · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
94 | 83k |
| 🥈 | docling Python library and CLI that converts PDFs, Office files and more into structured text 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
91 | 70k |
| 🥉 | opendataloader-pdf PDF parser that outputs Markdown, JSON with bounding boxes, and Tagged PDF Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
87 | 30k |
| #4 | markitdown Python library and CLI that converts office files, PDFs and audio to Markdown 🐳 Docker · MIT |
86 | 193k |
| #5 | agent-reach CLI that gives coding agents read access to login-gated and Chinese web platforms MIT · 📖 Docs ↗ |
85 | 107k |
| #6 | Crawl4AI Async Python web crawler that outputs LLM-ready Markdown and structured JSON 🐳 Docker · Apache-2.0 · 📖 Docs ↗ · 🌐 Site ↗ |
83 | 87k |
| #7 | Firecrawl Web API that turns sites and searches into markdown or JSON for LLMs 🐳 Docker · AGPL-3.0 · ▶️ Demo ↗ · 📖 Docs ↗ · 🌐 Site ↗ |
80 | 198k |
| #8 | labelme Desktop image annotation tool with polygon labeling and SAM-assisted masks GPL-3.0 · 📖 Docs ↗ · 🌐 Site ↗ |
79 | 16k |
| #9 | langextract Python library that extracts grounded structured data from text using LLMs 🐳 Docker · Apache-2.0 · ▶️ Demo ↗ · 📖 Docs ↗ |
78 | 39k |
| #10 | scrapegraph-ai Python library that builds LLM-driven scraping pipelines from a prompt and a URL 🐳 Docker · MIT · GPU optional · 📖 Docs ↗ · 🌐 Site ↗ |
78 | 32k |
Top 10 of 33. See all 33, ranked, with reviews →
💡 How to choose- Test on your hardest pages (JavaScript-heavy, tables, scans) before choosing.
- Check whether it runs fully self-hosted or needs the vendor's API for parts.
- Look at output formats: markdown, JSON with a schema, or both.
🔌 MCP servers
Model Context Protocol servers that give AI assistants tools and data. Coding-focused ones are on Best of Vibe Coding. 1 ranked · 📝 all reviews · 🌐 on archestack.com
servers leads with 93.
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | servers Reference MCP servers for files, git, fetch, memory and time custom license · 📖 Docs ↗ · 🌐 Site ↗ |
93 | 92k |
🧭 Go deeper: Glama ↗ grades thousands of MCP servers on security and quality.
💡 How to choose- Check what it can write or delete, not only what it reads.
- Prefer servers that scope credentials per tool or per user.
- See whether it runs locally (stdio) or as a remote server you have to secure.
📖 Learn
📖 Learning and cookbooks
Courses, cookbooks and guides with runnable code, for learning to build with AI. 65 ranked · 📝 all reviews · 🌐 on archestack.com
ai-engineering-from-scratch leads with 89, ahead of generative-ai-for-beginners (81) and rag_techniques (79).
| Rank | Project | Score | Stars |
|---|---|---|---|
| 🥇 | ai-engineering-from-scratch Hands-on AI engineering curriculum with 542 lessons across 20 phases MIT · 🌐 Site ↗ |
89 | 76k |
| 🥈 | generative-ai-for-beginners 21-lesson course on building generative AI apps in Python and TypeScript MIT |
81 | 123k |
| 🥉 | rag_techniques Collection of 42 runnable notebooks implementing RAG techniques custom license |
79 | 30k |
| #4 | cs249r_book Open textbook and course materials for machine learning systems engineering custom license · 📖 Docs ↗ · 🌐 Site ↗ |
79 | 30k |
| #5 | ai-agents-for-beginners 18-lesson course on building AI agents with Microsoft Agent Framework MIT · 📖 Docs ↗ |
78 | 79k |
| #6 | llms-from-scratch Code for the book on building a GPT-style LLM in PyTorch custom license · GPU optional · 🌐 Site ↗ |
77 | 108k |
| #7 | ai-for-beginners 12-week, 24-lesson AI curriculum with PyTorch and TensorFlow notebooks MIT · ▶️ Demo ↗ |
77 | 70k |
| #8 | agents-towards-production Jupyter tutorials for taking LLM agents from prototype to production custom license · GPU optional · 🌐 Site ↗ |
77 | 22k |
| #9 | awesome-llm-apps Collection of 100+ runnable LLM agent, RAG and voice app examples in Python Apache-2.0 · GPU optional · 🌐 Site ↗ |
76 | 144k |
| #10 | openai-cookbook Example notebooks and guides for building with the OpenAI API MIT · 📖 Docs ↗ · 🌐 Site ↗ |
76 | 77k |
Top 10 of 65. See all 65, ranked, with reviews →
💡 How to choose- Check the date: guides older than a year often use APIs that changed.
- Prefer material with runnable notebooks or repos over slides.
- Pick by level: curricula start from zero, cookbooks assume you already build.
🏅 How we rank
Every project gets a score out of 100 from the signals below. A signal we cannot measure yet is left out and its weight goes to the others, so nobody loses points for things we have not checked. The score sets the order; 🥇🥈🥉 mark the top three places in each table.
| Signal | Weight | What it looks at | Shown as (80+ · 50+ · 30+ · lower) | Status |
|---|---|---|---|---|
| ⭐ Adoption | 15% | GitHub stars, log-scaled, ranked half within the category and half across the list | widely used · popular · known · niche | live |
| 📈 Momentum | 10% | stars gained in the last 30 days, ranked within the category | surging · rising · steady · flat | soon |
| 🕒 Freshness | 15% | last commit (full marks under 90 days, zero at a year) and last release | active · recent · slowing · quiet | live |
| 🔧 Maintenance | 15% | issues closed vs opened, release cadence, commits in 90 days | healthy · fair · patchy · weak | live |
| ✅ Verified build | 30% | we start it in CI and wait for a healthy response | passes · untested · fails | soon |
| 📖 Docs | 5% | install section, environment docs, upgrade notes | thorough · good · thin · missing | soon |
| 🤖 Agent-ready | 5% | CLAUDE.md, AGENTS.md, skills, llms.txt | ready · partly · minimal · none | live |
| 🔒 Privacy | 5% | no telemetry or opt-out, works fully offline | private · mostly private · some telemetry · leaky | soon |
Penalties: no license −10. Coming: data must leave your host −10 · unpatched security advisory −20. Archived projects are removed. Paid placements never touch the score.
The weights are our judgment, not science. Think one is wrong? Tell us. Method changes: 2026-10-14: the star-growth anomaly penalty is dropped, since a star chart cannot tell a big launch from bought stars · 2026-10-13: Maintenance weighs 15% (was 10%) and Momentum 10% (was 15%), so steady upkeep counts more than a burst of stars · 2026-10-12: penalties we do not check yet are marked coming; only "no license" has ever been applied · 2026-10-09: Docker alone earns no points, medals mark places 1–3 instead of score bands.
🧠 How this works
- 🧠 On this list: open-source projects whose main job is AI, from apps you run to libraries you build with. Tools for coding with AI, starters and the stack you ship on live on the sister list, Best of Vibe Coding.
- 🔎 Found in curated lists, app stores and template galleries, then fact-checked on GitHub: stars, last commit, license, Docker files.
- ✍️ Written from the README, never copied from other lists: what it does, what it needs, strengths and weaknesses as claims you can check. Specs say
unknownrather than guess. - 🔄 Kept fresh: entries are rewritten when the README or the latest release changes; projects quiet for 12 months get marked stale, archived ones drop off.
- 🚫 No pay-to-rank. Sponsors and affiliate links, if any, are labeled and never touch the order.
📬 Your turn
➕ Add a project · 🛠️ Fix wrong data · 🚪 Opt out (done within 24 hours). Everything here is generated, so use the forms instead of editing files.
📚 Where candidates come fromThese lists, app stores and galleries (facts and links only, no text copied), plus community submissions: awesome-selfhosted · awesome-local-llm · awesome-homelab · self-hosting-guide · awesome-ai-agent-platforms · awesome-openclaw · umbrel-apps · runtipi-appstore · casaos-appstore · dokploy-templates · gh-search-self-hosted-ai · gh-search-ai-agents · gh-search-llm-apps · gh-search-ai-security · awesome-opensource-ai · awesome-llm-services · awesome-llmops · awesome-private-ai · awesome-local-llms · awesome-llm-webapps · 0xsojalsec-llms-local · inftyai-awesome-llmops · bradagi-awesome-cli-coding-agents · best-of-ml-python · awesome-generative-ai · awesome-ai-agents-e2b · awesome-mcp-servers · gh-search-llm-top · gh-search-ml-top · gh-search-agents-top · gh-search-mcp-top · field-image-comfyui · field-image-sd-zh · field-video-diffusion · field-video-generation · field-video-models · field-audio-scientific · field-audio-voice · field-audio-tts · field-audio-music · field-3d-gaussian-splatting · field-3d-scene-generation · field-science · field-science-bioinformatics · field-health · field-data-science · field-data-text2sql · field-trading-quant · field-trading-ai-finance · field-robotics-libraries · field-robotics-llm · field-robotics-embodied · field-research-deep-research · field-ocr · field-ocr-llm · gh-field-stable-diffusion · gh-field-video-generation · gh-field-text-to-speech · gh-field-medical-imaging · field-trading-systematic · field-trading-crypto-bots · field-trading-quant-ai · field-finance-fintech · field-finance-billing · field-health-medllms · field-health-ai4med · field-health-medical-imaging · field-commerce-platforms · field-commerce-recommenders · field-robotics-tooling · gh-field-3d-generation · gh-field-voice-conversion · gh-rising-new · gh-rising-half-year · gh-trending · repodir-recent.
📄 License
Data (data/, this README, reviews/) is CC BY 4.0; see LICENSE-DATA. Code is MIT; see LICENSE. Project names and descriptions belong to their owners.
Curated with care, powered by public GitHub data, funded by freelance gigs. Found a gem? Tips keep the servers happy, but rankings stay honest. 🚀
For freelance collabs or coffee-fueled conversations, say hello on LinkedIn.
Made with ☕ by Archestack · on the web at archestack.com · see also Best of Vibe Coding
Yorumlar (0)
Yorum birakmak icin giris yap.
Yorum birakSonuc bulunamadi