Catalog projects marked with #llm. Tags work as dedicated landing pages, so related tools are easier to find and connect.
This collection holds 55 projects with a combined 4,548,466 GitHub stars. Main languages: Python, TypeScript, Go.
Large language models stopped being research-lab magic — people now run them locally, fine-tune them on their data, and ship them inside products. The open source ecosystem around LLMs moves faster than any other: new tools land literally every week.
This collection covers inference engines, RAG and agent frameworks, API wrappers, and local runners. If you are building on top of language models, start here: these are proven projects under active development.
LLM Course is a learning roadmap for large language models covering fundamentals, fine-tuning, quantization, evaluation, notebooks, and engineering scenarios.
LobeHub is a platform for working with AI agents, knowledge, and collaborative task organization around large language models.
GPT4All is a project for running large language models locally on personal computers and inside applications.
OpenHands is an open AI software agent project with a CLI, local interface, SDK, and code-work scenarios.
Prompt Engineering Guide is an open learning resource for prompting, RAG, agents, and practical work with large language models.
LLaMA-Factory is a Python tool for tuning and training LLMs/VLMs with CLI, web UI, and support for popular models.
GPT Academic is a Chinese LLM interface focused on reading, translating, polishing, and analyzing academic material.
Unsloth provides tools and Studio for running and fine-tuning open models locally, with a focus on speed, memory, and practical LLM workflows.
RTK is a Rust proxy for development commands that reduces token usage when working with language models.
Open Interpreter is a lightweight coding agent for commands, files, and open LLMs.
AnythingLLM is an application for documents, agents, and multiple LLM providers in one interface.
Mem0 is a memory layer for AI agents and applications that need to store and search user context.
