AI Application / RAG Pipeline Builder
6.5k 2026-04-13
run-llama/rags
A Streamlit app to build and query custom Retrieval-Augmented Generation (RAG) pipelines over your data using natural language.
Core Features
Create RAG pipelines from data sources (local files, web pages) using natural language.
Configure RAG parameters (system prompt, top-k, summarization, LLM, embedding model) via a UI.
Interact with the generated RAG agent through a chatbot interface to query your data.
Supports various LLMs and embedding models, including OpenAI, Anthropic, Replicate, and local HuggingFace models.
Quick Start
poetry install --with dev && streamlit run 1_🏠_Home.pyDetailed Introduction
RAGs is an innovative Streamlit application designed to democratize the creation of Retrieval-Augmented Generation (RAG) pipelines. Inspired by OpenAI's GPTs, it allows users to build a custom 'ChatGPT over their data' by simply describing their task and data source in natural language. The project simplifies complex RAG configurations, offering a user-friendly interface to fine-tune parameters and interact with the generated AI agent, making advanced AI capabilities accessible for personalized data querying.