GURPREETKAURJETHRA/RAG-using-Llama3-Langchain-and-ChromaDB

Implements document-based question answering by embedding user documents into ChromaDB's vector store, then retrieving relevant chunks during inference to augment Llama3's context window. The system uses Langchain to orchestrate the retrieval pipeline and generation workflow, enabling accurate responses about custom documents like the EU AI Act without requiring model fine-tuning. Validated against real regulatory text to demonstrate RAG's effectiveness in grounding LLM outputs to specific document sources.

131 stars. No commits in the last 6 months.