• Picks
  • Discover
  • Alternatives
  • About

© 2026 Willet

  • Newsletter
  • About
  • FAQ
  • Terms
  • Privacy & Cookies
Join the Discord (opens in a new tab)
OpenBMB avatar

UltraRAG

MCP-based framework for building and evaluating RAG pipelines in YAML

  • Agent frameworks and tools
  • AI models and runtimes
View on GitHub
Website

What it is

UltraRAG splits RAG components such as retrievers and generators into separate MCP servers, then chains them with an MCP client using YAML files that support loops and conditional branches. It includes built-in benchmarks and evaluation workflows, plus a web UI with a visual pipeline builder, knowledge base management, and one-command conversion of a pipeline into a chat demo. Researchers and engineers use it to prototype, compare, and reproduce retrieval-augmented generation methods without writing much glue code.

Images

Facts

Stars
5,732 GitHub stars.
5,732

Forks
454
Commits
459
Contributors
23
Language
Python
License
Apache-2.0
Last commit
Oct 13, 2026
On GitHub since
January 2025
First release
October 2025
Platforms
web, Linux, self-hosted, library
Self-hostable
Yes
Tags
  • low-code
  • multimodal
  • llm-evaluation
  • knowledge-base
  • rag

Last synced from GitHub on Oct 20, 2026

The 10 best overlooked open-source projects, every Monday.

Share and more

Comments

No comments yet. Start the conversation.

Similar projects

  • ✨Data Intelligence Lab@HKU✨ avatar

    RAG-Anything

    Multimodal RAG framework for documents with images, tables, and equations

    23.9k
  • ✨Data Intelligence Lab@HKU✨ avatar

    LightRAG

    Graph-based retrieval-augmented generation library with a web UI

    40.2k