xiry.

AI · GitHub repo

OpenRAG

All-in-one local stack for building and running RAG applications

From the Xiry library · Updated 10 October 2026

Visit opensourceprojects.dev ↗

OpenRAG is an open-source, local-first platform designed to build and test Retrieval-Augmented Generation pipelines without relying on external cloud APIs. It bundles document parsing, vector storage, and local LLM execution into a single Docker Compose environment with a clean web interface.

What it's for

  • Run chat-over-documents locally without sending private files to external APIs
  • Prototype and test RAG pipeline components on a laptop before production deployment
  • Inspect document chunking, vector storage, and retrieval steps for educational or debugging purposes
  • Demo a functional document Q&A assistant to team members using a prebuilt UI

Features

  • One-command deployment using Docker Compose
  • Built-in web UI for document uploading and conversational querying
  • Local model execution powered by Ollama and nomic-embed-text
  • Vector storage and retrieval integration via ChromaDB
  • Transparency across the full pipeline from chunking to prompt synthesis

Pricing

Open source

Details

Type
GitHub repo
Category
AI
Topic
RAG Platforms
Open source
Yes
Platforms
Self-hosted, Linux, macOS, Windows, Web
For
Developers and AI engineers wanting a local, self-contained environment to build and test RAG workflows.
Alternative to
Dify, AnythingLLM, FastGPT
Tags
rag, langflow, ollama, chromadb, local ai, docker

Why it was saved

  • @GithubProjects, 9 March 2026: Replaces the usual patchwork of ~10 tools needed for a RAG system with a single stack for building and deploying RAG applications locally

Compare

Related

Updated