Incremental engine for long horizon agents 🌟 Star if you like it!
AI Open Source
Hand-picked AI open-source projects on GitHub worth following: LLMs, agents, RAG, local deployment, image and speech… 1051 projects so far.
Looking for projects you can deploy for free on Cloudflare, Vercel, Netlify and more? See Free Deploy Picks
All 1051🧠 Large Language Models 25🤖 AI Agents 165🧩 App Development Platforms 23📚 RAG & Knowledge Bases 48💬 Chat UIs & Clients 44🖥️ Local Inference & Deployment 85👨💻 AI Coding 97🛠️ Frameworks & SDKs 48🔌 MCP & Tool Ecosystem 36🎨 Image Generation 26🎬 Video Generation 17🎙️ Speech & Audio 45👁️ Computer Vision 27📄 Document Parsing & Scraping 25🗄️ Vector Databases 27🏋️ Training, Fine-Tuning & MLOps 74🧪 Evaluation, Observability & Safety 78🚪 API Gateways & Proxies 12🔎 Search & Deep Research 14⚙️ Deep Learning Libraries 56🏷️ Data & Annotation 33🔬 Industry & Science AI 34⚡ Productivity Tools 12
自动发现数据与标签中的问题。
面向多模态 AI 的嵌入式向量数据库。
面壁智能的高效端侧模型系列，小参数量、强能力，适合手机与边缘设备。
Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Codex, Gemini and more), by model, project, and task. npx codeburn
把书、长视频、播客等高价值内容蒸馏成可执行的 Agent Skills（Distill high-value content from books, long-form videos, podcasts, and more into executable Agent Skills）
Agent workspace built on Cloudflare Workers for creating documents, building apps, and running agents with your company’s context and systems.
🐍 Geometric Computer Vision Library for Spatial AI
An open-source framework for detecting, redacting, masking, and anonymizing sensitive data (PII) across text, images, and structured data. Supports NLP, pattern matching, and customizable pipelines.
Solution for long term memory for agent coding CLIs and to facilitate handoff between different agent vendors
实验追踪与模型管理 SDK。
Pocket Flow: 100-line LLM framework. Let Agents build Agents!
实时语音对话模型，全双工低延迟。
Very low latency speech to text, intent recognition, and text to speech, for building voice agents and interfaces
Build high-quality LLM apps - from prototyping, testing to production deployment and monitoring.
可视化、管理并改进计算机视觉数据集。
极速的现代分词器实现。
Point Cloud Library (PCL)
首个 DiT 视频生成模型，接近实时生成。
月之暗面的万亿参数 MoE 模型，擅长智能体与代码任务。
Multi-Agent Harness for Production AI
可视化测试与调试 MCP 服务器的官方工具。
NVIDIA 的通用模型推理服务器，支持多框架、多模型并发部署。
An Open-Source Asynchronous Coding Agent