Voyage AI
State-of-the-art embedding and reranking models for retrieval.
Acquired — still operating
acquired by MongoDB Feb 2025
About
Voyage AI builds frontier retrieval models: text embeddings, contextualized chunk embeddings, multimodal embeddings, and instruction-following rerankers that consistently rank near the top of the Retrieval Embedding Benchmark. Its Matryoshka-style embeddings let developers truncate dimensions without re-embedding a corpus.
MongoDB acquired Voyage AI in February 2025 and folded the models into Atlas as its retrieval stack, powering vector search, hybrid search, and automated embeddings directly inside the database. Voyage models now ship as MongoDB's unified data intelligence layer for production AI applications.
What it can do
- Text and multimodal embeddings
- Contextualized chunk embeddings
- Instruction-following rerankers
- Native MongoDB Atlas integration
Setup
Sign up at dash.voyageai.com or use through MongoDB Atlas; REST API with API key.
Prerequisites
API key; a MongoDB Atlas cluster if using the integrated path.
Use cases
Works with
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 105; Trending score 88/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.