README.md

metadata

title: RAG Generation Service
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: docker
pinned: false
license: mit

RAG Generation Service

This is a Retrieval-Augmented Generation (RAG) service that answers questions based on provided context.

How to use

1. Enter your question in the "Query" field
2. Paste relevant documents or context in the "Context" field
3. Click submit to get an AI-generated answer based on your context

Features

• Uses state-of-the-art language models via Hugging Face Inference API
• Supports multiple model providers
• Clean, intuitive interface
• Example queries to get started

Configuration

This Space requires a HF_TOKEN environment variable to be set with your Hugging Face access token.

Model Support

By default, this uses meta-llama/Meta-Llama-3-8B-Instruct, but you can configure different models via environment variables.