voyage-4-lite
Voyage 4 Lite is the smaller hosted encoder, designed to reduce inference cost.
Model design
Lite uses the same embedding space as the other Voyage 4 models. It can encode a query against documents embedded with Large. This compatibility still requires matching output dimensions.Voyage 4 model family
Practical use
Practical guidance. Use a smaller query encoder when search traffic is frequent and the document index changes less often. Compare short, ambiguous queries separately from long questions; average recall can hide differences between these groups.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Launch date | |
| Access | API |
| Atlas lifecycle | ActiveMongoDB Voyage AI model lifecycle: Atlas scope and release history |
| Input limit | 32,000 tokensMongoDB Voyage AI model specifications |
| Output dimensions | 1,024 default; 256, 512 or 2,048 optionalMongoDB Voyage AI model specifications |
Related models
General text embeddings compatible with the other Voyage 4 models.
Launch date API- voyage-4-large
Voyage AI · Voyage 4
Mixture-of-experts text embeddings with a shared Voyage 4 vector space.
Launch date API - voyage-4-nano
Voyage AI · Voyage 4
Downloadable multilingual embedding model with bidirectional attention.
Launch date Downloadable