The RAG Framework Directory You Didn't Know You Needed

If you've been building with RAG (Retrieval-Augmented Generation) recently, you've probably noticed the same thing I have: the ecosystem is exploding. There's LangChain, LlamaIndex, Chroma, Weaviate, and a dozen other frameworks, each with different strengths, trade-offs, and learning curves. Picking the right one for your project can feel like trying to find a specific needle in a rapidly growing haystack.

That's exactly why RAGHub exists. It's a community-driven directory that maps out the RAG landscape, so you don't have to waste hours comparing docs and repos on your own. It's not a framework itself—it's a curated guide to help you decide which framework to use for your use case.

What It Does

RAGHub is essentially a living, browsable index of RAG-related tools, libraries, and frameworks. Each entry includes a brief description, a link to the original repo or project page, and a rough categorization (like "orchestration," "embedding," "storage," or "evaluation"). The goal is simple: give you a quick, honest overview of what's out there, so you can make an informed choice without drowning in options.

The repo itself is just a markdown file—no bloat, no build steps. You can read it directly on GitHub, or contribute your own additions if you find something missing.

Why It's Cool

The best thing about RAGHub is how practical it is. Here's what stands out:

• Curated, not exhaustive. It doesn't try to list every single library that touches RAG. Instead, it focuses on the ones that actually matter—frameworks people are using in production or on popular projects.
• Community-driven updates. Because it's open source, anyone can submit a PR to add a new tool, fix a broken link, or update a description. This means it stays relevant as the ecosystem evolves.
• Quick scanability. The directory is organized by category, so you can go straight to "vector databases" or "document loaders" without scrolling through irrelevant entries. Perfect for when you need a quick answer during a coding session.
• No vendor bias. Unlike some sponsored lists or blog posts, this is just a straightforward collection from developers who've actually used these tools. No fluff, no affiliate links.

How to Try It

You don't need to install anything. Just head over to the repository:

https://github.com/Andrew-Jang/RAGHub

Open up the README.md (or the main list file—check the repo structure), and you'll see the directory right there. Scroll through, find what you need, and click through to the framework's repo or docs. That's it.

If you want to contribute, it's as simple as forking the repo, editing the markdown file, and opening a pull request. No complex CI or formatting rules—just good old-fashioned markdown.

Final Thoughts

RAGHub won't build your app for you, but it will save you from the "which framework should I use" rabbit hole that every RAG developer inevitably falls into. Whether you're prototyping a simple Q&A bot or architecting a production retrieval pipeline, having a quick reference like this is genuinely useful.

If you've been frustrated by the "framework of the week" pace in the LLM world, give it a look. It's the kind of resource you'll find yourself coming back to more often than you'd expect.

Brought to you by @githubprojects