Voyage Rerank 3 and Rerank 3 Lite are now available on AI Gateway

Rerank 3 and Rerank 3 Lite from Voyage AI by MongoDB are now available on AI Gateway.

The models reorder search results by relevance before passing documents to a language model. The series improves retrieval quality across domains, with the largest gains on long documents and code.

Rerank 3 focuses on accuracy, while Rerank 3 Lite is optimized for latency and cost. Both handle up to 32K tokens per query-document pair and follow instructions in the query to guide ranking.

To use Rerank 3, set the model to voyage/rerank-3:

1
import { rerank } from 'ai';
2

3
const { ranking } = await rerank({
4
  model: 'voyage/rerank-3',
5
  query: 'How does a vector database find similar documents?',
6
  documents: [
7
    'A load balancer distributes traffic across servers.',
8
    'A vector database searches embeddings to find similar documents.',
9
  ],
10
  topN: 1,
11
});

For Rerank 3 Lite, change the model to voyage/rerank-3-lite.

Try Voyage Rerank 3 in the AI Gateway model playground, or explore the model catalog.

AI Gateway provides one API for calling models, tracking usage and cost, and configuring routing, retries, and failover. You can use an AI Gateway API key or bring your own provider key.