Wispr Flow - Don’t type, just speak
Mini toolsVoyage 4
A defining feature of the Voyage 4 series is a shared embedding space: all four models (voyage-4-large, voyage-4, voyage-4-lite, voyage-4-nano) produce compatible vectors. This enables asymmetric retrieval where documents are embedded once with a larger model and queries are served using a smaller, faster model without re-indexing. voyage-4 approaches the retrieval quality of voyage-3-large while maintaining efficiency of a mid-sized model. Evaluated across the 29-dataset RTEB benchmark, it outperforms OpenAI text-embedding-3-large by a wide margin. Available via the Voyage API and through MongoDB Atlas.
Overview
Text embedding model optimized for general-purpose and multilingual retrieval quality. Supports a 32,000-token context window and configurable output dimensions (256, 512, 1024, 2048) via Matryoshka Representation Learning. Part of the Voyage 4 series, which shares a common embedding space enabling asymmetric retrieval across models in the family.
About Voyage AI
Voyage AI provides best-in-class embedding models and rerankers for search and retrieval over unstructured data, used to power retrieval-augmented generation (RAG) and AI applications. It offers general-purpose, domain-specific (finance, legal, code) and company-specific fine-tuned models. Founded in 2023 and based in Palo Alto, the company was acquired by MongoDB, Inc. in February 2025 and now operates as a MongoDB subsidiary.
View Company Profile
