Shubhamsaboo/awesome-llm-apps
100+ AI Agent & RAG apps you can actually run — clone, customize, ship.
Awesome LLM Apps by Shubhamsaboo is the most popular collection of runnable AI Agent and RAG apps on GitHub, with 110,000 stars and 100+ self-contained Python projects you can clone, customize, and ship. If you want working code instead of slide decks, start here.
Our Review
Awesome LLM Apps is the most popular collection of runnable AI Agent and RAG example apps on GitHub -- 110,000 stars and 16,400 forks as of May 2026. Maintained by Shubham Saboo (Shubhamsaboo on GitHub, founder of the Unwind AI newsletter) under the Apache 2.0 license, it ships 100+ self-contained Python projects you can clone, customize, and run today.
What Awesome LLM Apps does:
- •100+ runnable example apps each in its own folder with a complete README, requirements.txt, and ready-to-run Python code.
- •AI Agent + RAG focus agents that book travel, agents that scrape and summarize, RAG over PDFs, RAG over websites, multi-agent collaboration, autonomous research.
- •Real LLM integrations examples using OpenAI, Anthropic Claude, Google Gemini, Groq, Ollama, and Hugging Face local models.
- •Framework variety LangChain, LlamaIndex, CrewAI, AutoGen, OpenAI Assistants, plus from-scratch agents using only the OpenAI SDK.
- •Self-contained per app no monorepo coupling; you can fork one folder and ignore the rest.
- •Apache 2.0 licensed commercial use allowed with attribution, no copyleft requirement.
- •Active maintenance new apps added weekly as the AI agent ecosystem evolves.
Awesome LLM Apps ecosystem:
- •Unwind AI newsletter the maintainer's free weekly newsletter at theunwindai.com that covers practical LLM application patterns.
- •Per-app READMEs detailed setup instructions, screenshots, and explanation per example.
- •Community PRs accepts contributions for new app categories and framework examples.
Getting started:
Browse github.com/Shubhamsaboo/awesome-llm-apps and pick an app folder that matches your use case. Each folder has its own README. Typical setup: cd app-folder && pip install -r requirements.txt, set your API key in .env, then python app.py or streamlit run app.py. Most apps run in under five minutes from clone to working demo.
Limitations:
This is a teaching collection, not a production framework, so expect rough edges, hardcoded paths, and example-quality error handling. Apps may use outdated framework versions and need manual updates to track upstream library changes. Some apps require multiple paid API keys (OpenAI plus Tavily plus Pinecone, for instance), which can add up across experiments. The collection grows fast, so finding the right example for a specific niche requires browsing or searching the README index.
Our Verdict
Awesome LLM Apps in 2026 is the most-starred collection of runnable AI Agent and RAG example code on GitHub. 110,000 stars, 16,400 forks, Apache 2.0 license, and 100+ self-contained Python apps maintained by Shubhamsaboo (Shubham Saboo).
For Vibe Builders, this is the fastest way from "I want to build a thing that does X with AI" to working code. Clone one folder, install dependencies, run it. Each app is small enough to read end-to-end in 30 minutes, making it ideal for learning by example rather than reading docs.
For Developers, the collection is a reference library across every major agent framework (LangChain, LlamaIndex, CrewAI, AutoGen, OpenAI Assistants). You can compare implementations of the same use case across frameworks and pick the one that fits your stack.
Skip Awesome LLM Apps if you need production-grade code with full error handling and tests -- these are example apps, not production templates. Skip if you only need an OpenAI Cookbook-style snippet library -- the openai/openai-cookbook repo is more focused on raw API patterns. For curated AI Agent platforms (instead of examples), Dify or AutoGPT are full products rather than starter code.
Frequently Asked Questions
What is Awesome LLM Apps?
Awesome LLM Apps is a curated collection of 100+ runnable AI Agent and RAG example apps on GitHub, maintained by Shubham Saboo (Shubhamsaboo on GitHub). Each app is a self-contained Python project with its own README and requirements file, covering frameworks like LangChain, LlamaIndex, CrewAI, and AutoGen. As of May 2026 it has 110,000 GitHub stars under Apache 2.0.
What kinds of apps are in the collection?
The collection covers AI agents that book travel and scrape data, RAG apps that answer questions over PDFs and websites, multi-agent collaboration patterns, autonomous research bots, and from-scratch agents using only the OpenAI SDK. Frameworks represented include LangChain, LlamaIndex, CrewAI, AutoGen, and OpenAI Assistants for a broad survey of approaches.
How do I use Awesome LLM Apps?
Browse the repository on GitHub, pick an app folder that matches your use case, and follow that folder's README. Typical setup is cd app-folder, pip install -r requirements.txt, set your API keys in .env, then run python app.py or streamlit run app.py. Most apps go from clone to working demo in under five minutes.
Is Awesome LLM Apps free to use commercially?
Yes. Awesome LLM Apps is licensed under Apache 2.0, which permits commercial use, modification, and distribution with attribution. The license has no copyleft requirement, so you can fork an app and use it in a closed-source commercial product. You need to keep the original license and copyright notices in any forked code.
Who maintains Awesome LLM Apps?
Awesome LLM Apps is maintained by Shubham Saboo (Shubhamsaboo on GitHub), the founder of the Unwind AI newsletter at theunwindai.com. He curates the collection, accepts community pull requests for new apps, and ships updates weekly to track the evolving AI agent ecosystem and new framework releases.
What is awesome-llm-apps?
Awesome LLM Apps by Shubhamsaboo is the most popular collection of runnable AI Agent and RAG apps on GitHub, with 110,000 stars and 100+ self-contained Python projects you can clone, customize, and ship. If you want working code instead of slide decks, start here.
What license does awesome-llm-apps use?
awesome-llm-apps uses the Apache-2.0 license.
What are alternatives to awesome-llm-apps?
Explore related tools and alternatives on My AI Guide.
Open source & community-verified
Apache-2.0 licensed: free to use in any project, no strings attached. 111,248 developers have starred this, meaning the community has reviewed and trusted it.
Reviewed by My AI Guide for relevance, quality, and active maintenance before listing.
Topics