GitHub - samrudhau/Multimodal-GraphRAG: A multimodal RAG system that builds a unified knowledge graph from both audio transcripts and PDF documents. LLMs extract entities, relationships, and chunk semantics. Neo4j stores the graph and vector embeddings. A FastAPI service answers questions by combining vector search, graph expansion, and LLM reasoning with source citation.

Name		Name	Last commit message	Last commit date
data		data
src		src
.dockerignore		.dockerignore
Dockerfile		Dockerfile
docker-compose.yml		docker-compose.yml
requirements.txt		requirements.txt

About

A multimodal RAG system that builds a unified knowledge graph from both audio transcripts and PDF documents. LLMs extract entities, relationships, and chunk semantics. Neo4j stores the graph and vector embeddings. A FastAPI service answers questions by combining vector search, graph expansion, and LLM reasoning with source citation.

docker fastapi neo4j pymupdf qwen2-5 rag-pipeline whisper

Activity

0 stars

0 watching

0 forks

Report repository