Best-in-class embedding and reranking models, now part of MongoDB, purpose-built for RAG.
Voyage AI builds domain-tuned embedding and reranking models that consistently top retrieval benchmarks โ legal, code, finance, and multilingual. Acquired by MongoDB in 2024, it's the default upgrade teams make when generic embeddings aren't cutting it for RAG.
Who it's for: Developers who need top-tier retrieval quality for RAG and search.
Domain-tuned for code, law, finance.
Big precision gains on top of vectors.
Strong across languages.
Drop-in for any pipeline.
If retrieval quality is your bottleneck, Voyage's embeddings + reranker is the highest-leverage upgrade you can make. Pair with Qdrant or Pinecone.