metadata

title: RAG Document Processing
emoji: 📄
colorFrom: blue
colorTo: green
sdk: docker
app_port: 7860
pinned: false
models:
  - sentence-transformers/all-MiniLM-L6-v2
  - facebook/bart-large-cnn
license: apache-2.0

RAG Document Processing

This Hugging Face Space contains a RAG (Retrieval Augmented Generation) system for processing documents, built entirely with Hugging Face tools and models.

Features

Automated processing of text files
OCR and document processing with HF models
Vector storage with FAISS
Embeddings with sentence-transformers
Simple query interface with Gradio

Implementation

This is a Hugging Face-native implementation of the original cloud-based RAG system. It uses:

Sentence Transformers for embeddings
FAISS for vector search
Transformers models for document processing
Persistent storage for data
A100 GPU acceleration (for pro users)

Usage

Upload documents to be processed
The system automatically processes and indexes them
Query the processed documents through the interface

See the CLAUDE_HF.md file for implementation details.