LAUNCHMongoDB 8.3 is built for the sub-100ms retrieval & zero downtime AI demands. Read blog >
AI DATAStop fighting your data layer. Get the memory & retrieval agents need to scale. Read blog >

Getting Started with Voyage AI

Follow our quick start guides to access state-of-the-art Voyage AI embedding and reranking models through a simple API call.

Get Started

Generate your first vector embedding

number_1

Create an API key

Model API keys are your secure credentials for authenticating with the Atlas Embedding and Reranking API Service, managed within the Atlas console.

Create an API key
number_2

Choose an embedding model

Voyage AI offers embedding models ranging from general-purpose, multimodal, contextualized chunk, and domain-specific models.

Learn more
number_3

Start building

Get started quickly with Python or TypeScript SDKs and leverage the best-performing models to power your semantic search, RAG, and agentic applications.

Follow the quick start guide

Explore more resources

mdb_document_model

Check out the docs

The Voyage AI documentation will introduce you to embedding models and rerankers and help you get started with Voyage AI models.

Read the docs
general_content_play

Watch a session on AI-powered retrieval

Examine techniques—embedding models, rerankers, query decomposition, chunking, and more—for enhancing retrieval performance.

Watch the video
general_features_insight

Learn more about Voyage AI

MongoDB’s Frank Liu, Staff Product Manager, and Richmond Alake, Staff Developer Advocate, break down powerful AI capabilities easily accessible to MongoDB developers.

Watch the video

Get started today with the Voyage AI API

Voyage AI, part of MongoDB, builds leading embedding models and rerankers. Access the new Atlas Embedding and Reranking API Service to start building high-accuracy AI applications instantly.
Try FreeContact sales
ADVANCED AI MODELS
  • High accuracy
  • Low dimensionality
  • Low-latency
  • Shared embedding space
  • Multimodal support