跳到正文
Vercel Blog· Zachary Chen·· 6 天前精選AI 評分61

Voyage Rerank 3 與 Rerank 3 Lite 於 AI Gateway 上可使用

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

AI 導讀

Voyage Rerank 3 與 Voyage Rerank 3 Lite 現已可在 Vercel 的 AI Gateway 直接使用,兩者皆支援 32K token 的查詢-文件對,並可依需求調整準確度或效能。

推薦理由

模型提升長文件與程式碼檢索品質,使用者可依需求選擇準確度或效能;同時在 AI Gateway 上一鍵使用,降低部署門檻,並提供完整 API 追蹤與成本管理。

正文 · 原文

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:

import { rerank } from 'ai';

const { ranking } = await rerank({

model: 'voyage/rerank-3',

query: 'How does a vector database find similar documents?',

documents: [

'A load balancer distributes traffic across servers.',

'A vector database searches embeddings to find similar documents.',

],

topN: 1,

});

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.

來源:Vercel Blog · vercel.com