Skip to content

AutoRAG-Research

Release Build status Commit activity License

Automate your RAG research with reproducible benchmarks.

What is AutoRAG-Research?

A Python framework for:

  • Running RAG benchmarks on standard datasets
  • Evaluating retrieval and generation pipelines
  • Comparing algorithms with reproducible metrics

Quick Start

pip install autorag-research
docker-compose up -d
autorag-research data restore beir scifact_openai-small
autorag-research run --db-name=beir_scifact_test_openai_small

Choose Your Path

I want to... Go to
Run text retrieval benchmarks Text Retrieval Tutorial
Run full RAG with generation Text RAG Tutorial
Work with visual documents Multimodal Tutorial
Use my own dataset Custom Dataset Tutorial
Test my own pipeline Custom Pipeline Tutorial
Create my own metric Custom Metric Tutorial

Documentation