Comfy-Org/ComfyUI
ComfyUI
ComfyUI is a node-based interface for creating images, video, and other media with fine control over models and parameters.
Open Source: #ai
Catalog projects marked with #ai. Tags work as dedicated landing pages, so related tools are easier to find and connect.
This collection holds 142 projects with a combined 11,881,711 GitHub stars. Main languages: Python, TypeScript, Rust.
About the AI projects here
Artificial intelligence is the fastest-growing corner of open source: the last couple of years brought thousands of GitHub tools around neural networks, from ready-made assistants to low-level inference engines. This page collects the ones genuinely worth opening: alive communities, clear docs, practical value.
Inside you will find frameworks for working with language models, agent systems, local-inference tools, and learning resources. Every project card shows stars, the license, and a human description so you can judge fit quickly.
Repositories
Found: 142
ggml-org/llama.cpp
llama.cpp
llama.cpp runs large language models locally, from command-line inference to an OpenAI-compatible API server.
Shubhamsaboo/awesome-llm-apps
Awesome LLM Apps
Awesome LLM Apps collects runnable language-model application examples: agents, RAG, voice scenarios, and MCP integrations.
microsoft/generative-ai-for-beginners
Generative AI for Beginners
Generative AI for Beginners is Microsoft’s 21-lesson course on building generative AI applications with Python and TypeScript.
github/spec-kit
Spec Kit
Spec Kit is GitHub’s toolkit for spec-driven development: scenarios and plans first, then tasks and implementation with an AI assistant.
msitarzewski/agency-agents
Agency Agents
Agency Agents is a set of specialized AI agents as Markdown files for Claude Code, Cursor, Codex, and other tools.
garrytan/gstack
gstack
gstack is a set of Claude Code skills and commands that turns AI-assisted development into roles, reviews, browser QA, and shipping steps.
google-gemini/gemini-cli
Gemini CLI
Gemini CLI is Google’s open-source AI agent for the terminal, with Gemini access, file operations, shell commands, web fetching, and MCP integrations.
deepseek-ai/DeepSeek-V3
DeepSeek-V3
DeepSeek-V3 is DeepSeek’s open model repository with a MoE architecture, 671B total parameters, 37B activated parameters, and materials for running and studying it.
openai/whisper
Whisper
Whisper is OpenAI’s model and Python package for speech recognition, speech translation into English, and language identification in audio files.
browser-use/browser-use
Browser Use
Browser Use is a Python tool for browser AI agents: the model gets a browser action space, profile, domain limits, and a task execution loop.
rasbt/LLMs-from-scratch
LLMs from scratch
LLMs from scratch is Sebastian Raschka’s educational repository for implementing a GPT-like language model step by step in PyTorch.