Skip to content
Vercel April 2026 security incident
Read the bulletin
All Models

Voyage 4 Lite

voyage/voyage-4-lite

Voyage 4 Lite is the lightweight Voyage 4 model. Voyage AI reports it approaches voyage-3.5 retrieval accuracy with fewer parameters, shares one embedding space with `voyage-4-large` and `voyage-4`, and supports a context window of 32K tokens with Matryoshka dimensions and quantization like the rest of the family.

index.ts
import { embed } from 'ai';


const result = await embed({
  model: 'voyage/voyage-4-lite',
  value: 'Sunny day at the beach',
})

What To Consider When Choosing a Provider

  • Zero Data Retention

    AI Gateway does not currently support Zero Data Retention for this model. See the documentation for models that support ZDR.

    Authentication

    AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.

Use Voyage 4 Lite for queries when voyage-4-large already holds your document vectors, or as a budget option when both sides use the same model and your accuracy targets match Voyage AI's voyage-3.5 positioning. Plan a full re-embed when moving into Voyage 4, and test on a sample before indexing the full corpus.

When to Use Voyage 4 Lite

Best For

  • High query traffic:

    Pair Voyage 4 Lite queries with voyage-4-large document embeddings to keep per-query cost low without re-indexing

  • Cost-sensitive symmetric indexing:

    Voyage 4 Lite on both sides when voyage-3.5-level retrieval accuracy is sufficient and per-token cost drives the decision

  • Early production and prototypes:

    Iterate cheaply before upgrading query-side models once traffic patterns stabilize

  • Batch jobs:

    Large-corpus indexing runs where per-token cost compounds across millions of requests

Consider Alternatives When

  • Higher published average scores:

    Use voyage-4-large or voyage-4 when retrieval accuracy matters more than per-token cost

  • Code-only corpora:

    Use voyage-code-3 for repositories where source code is the primary content type

  • Multimodal embeddings:

    Use a model with native image inputs when you need to embed diagrams, screenshots, or mixed-format documents

Conclusion

Pick Voyage 4 Lite when your embedding bill scales with request volume and you need Voyage 4 generation quality at the tightest possible price point. Route through AI Gateway to swap between Voyage 4 tiers without changing your integration.

FAQ

voyage-4 is the mid-sized Voyage 4 model; Voyage AI reports it approaches voyage-3-large quality. Voyage 4 Lite uses fewer parameters; Voyage AI reports it approaches voyage-3.5 accuracy. Both share the same embedding space as voyage-4-large.

Voyage 4 Lite is a Voyage 4 model with a shared embedding space and updated training. Voyage AI positions it near voyage-3.5 accuracy with fewer parameters. Moving from Voyage 3.x requires re-embedding.

32K tokens. Size chunks so single requests stay under this limit.

Yes. Voyage AI targets production for voyage-4-lite, including asymmetric setups with voyage-4-large document embeddings when query cost matters.

Add your Voyage AI API key in AI Gateway settings, then send embedding requests through AI Gateway. AI Gateway authenticates requests and records usage.

Move when you need higher query-side accuracy. If documents stay on voyage-4-large, you can upgrade query embeddings to voyage-4 or voyage-4-large without re-vectorizing documents in that asymmetric setup.

Yes. Voyage 3 and Voyage 4 use different embedding spaces, so you re-embed when you change generations.