# Retrieval

## Embedding Models

Power enterprise search and question answering with NVIDIA Nemotron RAG models by embedding text and scanned documents, including PDFs and images, for fast, multilingual, multimodal data retrieval.

- [llama-nemotron-embed-vl-1b-v2](/nvidia/llama-nemotron-embed-vl-1b-v2.md)

## Reranking Models

Improve information retrieval accuracy with world-class NVIDIA Nemotron RAG models by reranking retrieved enterprise data to improve answer relevancy.

- [llama-nemotron-rerank-vl-1b-v2](/nvidia/llama-nemotron-rerank-vl-1b-v2.md)

## Extraction Models

Accelerate large-scale extraction from massive collections of multimodal data - text, images, and complex documents - with NVIDIA Nemotron RAG models for rapid, context-aware insights across your enterprise.

- [nemoretriever-parse](/nvidia/nemoretriever-parse.md)

## Explore NVIDIA Blueprints

Accelerate AI application development with ready-to-use workflows powered by NVIDIA NIM and NeMo microservices for RAG, AI agents, video search and summarization, and more.

- [Build an Enterprise RAG Pipeline Blueprint](/nvidia/build-an-enterprise-rag-pipeline.md)
- [Build an AI Virtual Assistant](/nvidia/ai-virtual-assistant-for-customer-service.md)
- [Build a Video Search and Summarization (VSS) Agent](/nvidia/video-search-and-summarization.md)
- [NVIDIA AI-Q Blueprint for intelligent agents](/nvidia/aiq.md)
- [Code Documentation for Software Development](/crewai/code-documentation-for-software-development.md)