A lightweight, fast cross-platform AI assistant supporting web, desktop and mobile.
AI Open Source
Hand-picked AI open-source projects on GitHub worth following: LLMs, agents, RAG, local deployment, image and speech… 1045 projects so far.
Looking for projects you can deploy for free on Cloudflare, Vercel, Netlify and more? See Free Deploy Picks
All 1045🧠 Large Language Models 24🤖 AI Agents 162🧩 App Development Platforms 22📚 RAG & Knowledge Bases 48💬 Chat UIs & Clients 40🖥️ Local Inference & Deployment 88👨💻 AI Coding 97🛠️ Frameworks & SDKs 47🔌 MCP & Tool Ecosystem 36🎨 Image Generation 26🎬 Video Generation 17🎙️ Speech & Audio 45👁️ Computer Vision 26📄 Document Parsing & Scraping 25🗄️ Vector Databases 25🏋️ Training, Fine-Tuning & MLOps 71🧪 Evaluation, Observability & Safety 79🚪 API Gateways & Proxies 12🔎 Search & Deep Research 15⚙️ Deep Learning Libraries 58🏷️ Data & Annotation 32🔬 Industry & Science AI 34⚡ Productivity Tools 16
A modern open-source AI chat framework supporting multiple models, plugins and knowledge bases, with one-click deployment.
High-quality conversion of PDFs and other documents into Markdown / JSON, aimed at LLM corpora.
Fine-tune LLMs twice as fast with 70% less VRAM, so even a single GPU can train.
Unified, efficient fine-tuning of 100+ LLMs with a no-code web UI, the most popular fine-tuning tool in China.
Gives multiple GPTs the roles and processes of a software company: one sentence of requirements produces a complete project.
An autonomous coding agent inside VS Code that reads and writes files and runs commands, with built-in human confirmation.
IBM's open-source document parsing tool that converts PDF and Office files into structured data.
The classic Python machine-learning library: simple and efficient.
An all-in-one AI app: chat with documents, agents and vector databases, available as a desktop app and Docker.
A TTS tool that clones a voice and synthesizes speech from just a few seconds to a minute of audio.
The de facto standard implementation of the YOLO series: detection, segmentation, pose estimation and tracking in one place.
An SDK and gateway for calling 100+ models in the OpenAI format, with cost tracking and load balancing.
A role-playing multi-agent orchestration framework that lets agents collaborate like a team.
Build LLM flows and agents by drag and drop, one of the fastest low-code ways to get started.
Meta's "Segment Anything" model, a promptable foundation model for image segmentation.
SDXL image-generation software focused on prompts, ready to use after installing, with a Midjourney-like experience.
A desktop AI client supporting multiple model providers, with a built-in knowledge base and MCP.
A data framework for LLM applications that connects private data and builds RAG and agents.
An AI pair-programming tool in the terminal that edits your local Git repository directly and commits automatically.
A cloud-native, high-performance vector database for billion-scale vector search.
A unified AI compute engine for distributed training, tuning, inference and serving.
Build stateful, long-running multi-agent applications as graphs.
A cross-platform AI desktop client supporting mainstream models such as ChatGPT, Claude and Gemini.