NeMo Retriever Documentation
Initializing search
NVIDIA/NeMo-Retriever
NeMo Retriever Documentation
NVIDIA/NeMo-Retriever
1. Introduction
1. Introduction
NeMo Retriever Library Overview
Key concepts
Agentic retrieval (concept)
Release notes
2. Get started
2. Get started
About this section
Pre-Requisites & Support Matrix
Quickstart: NeMo Retriever Library
Authentication and API keys
3. Deployment options
3. Deployment options
Compare deployment options
OpenShift deployment (Helm)
Helm chart (Kubernetes)
Docker service image
4. Core workflows
4. Core workflows
Workflow: Ingest documents into a searchable VDB collection
Workflow: Audio & video ingestion
Workflow: Agentic retrieval
5. Multimodal extraction
6. Embedding, indexing & storage
6. Embedding, indexing & storage
Multimodal embeddings (VLM)
Vector databases
7. Deployment & operations
7. Deployment & operations
Ray and distributed ingest
Batch resource sizing
8. Customize & extend
8. Customize & extend
Customize & extend
9. Integrations & ecosystem
9. Integrations & ecosystem
Starter kits
10. Evaluation & benchmarks
10. Evaluation & benchmarks
Evaluate on your own documents
11. Reference
11. Reference
API guide
CLI reference
Quickstart: retriever CLI
Collection management Python API
Environment variables
Metadata reference
12. Support & community
12. Support & community
Troubleshooting
FAQ
Contributing
Additional resources
Additional resources
OSS licences
404 - Not found
Back to top