You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

⚡ Lumina RAG • Universal Document Intelligence

A state-of-the-art multi-document research assistant with Hybrid Retrieval, Cross-Encoder Reranking, and contextual LLM synthesis.

Hybrid Retrieval: Combines Dense Semantic Search (FAISS) with Sparse Lexical Search (BM25).
Cross-Encoder Reranking: Re-evaluates top candidates with ms-marco-MiniLM-L-6-v2 to suppress false positives.
Exact Page-Level Citations: Every claim is cited with the original document name, page number, and previewable text snippet.
Contextual Query Rewriter: Resolves conversational pronouns across chat history.
Multi-Document Support: Ingest and index multiple PDF documents simultaneously.

Under your Hugging Face Space Settings -> Variables and secrets, add:

Secret Name	Description
`OPENROUTER_API_KEY`	Your OpenRouter API Key (sk-or-v1-...)

pip install -r requirements.txt
python -m streamlit run app.py

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support