automation

AutoRAG: Automate RAG Evaluation

Name: AutoRAG: Automate RAG Evaluation
Author: Marker-Inc-Korea

Optimize Retrieval-Augmented Generation with AutoRAG, an open-source framework for startup founders, with 4.9k+ GitHub stars.

4,852 stars405 forksPythonQuality 8/10Updated 7/2/2026100% free · open source

What it does

AutoRAG optimizes Retrieval-Augmented Generation by automating the evaluation and optimization process with AutoML-style automation, allowing founders to fine-tune their language generation models efficiently

Install / run

git clone https://github.com/Marker-Inc-Korea/AutoRAG.git

When to use it

•When you need to improve the performance of your language generation models
•When you want to automate the process of evaluating and optimizing Retrieval-Augmented Generation models
•When you need to reduce the time and effort required to fine-tune your language generation models

Quick start

1Navigate to the cloned repository: cd AutoRAG
2Create a new environment: python -m venv rag-env
3Activate the environment: source rag-env/bin/activate
4Install required packages: pip install -r requirements.txt
5Run the example script: python examples/example.py

Ready-to-paste prompt

python scripts/train.py --model_name_or_path t5-base --dataset_name wiki_text --do_train --do_eval --train_batch_size 16 --eval_batch_size 64

Heads up: Make sure you have the correct Python version (3.8 or later) and the required packages installed, as specified in the requirements.txt file, to avoid compatibility issues

Saves to your device