We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
mongodb/rerank-2.5Runs on your balance at the prices above. New accounts start with free usage.
Settings
VoyageAI by MongoDB: rerank-2.5 is a reranking model from VoyageAI by MongoDB. It belongs to the rerank category, which means its job is to take an existing set of candidate results and reorder them by relevance rather than to generate new text from scratch. It is a text model, and its output is a rerank result.
The model accepts text as input and produces reranked output. It has a context length of 32,000 tokens and a maximum output of 28,800. Because it is a reranker, it does not write prose, answer questions or hold a conversation. Instead, it scores and orders passages relative to a query, which is the step that typically sits between an initial retrieval pass and the final set of documents handed to a downstream model or shown to a user.
That shape makes it useful wherever retrieval quality matters. A common pattern is a two stage search pipeline: a fast vector or keyword search returns a broad candidate list, and rerank-2.5 reorders those candidates so the most relevant items land at the top. This applies to retrieval augmented generation, where better ordering means the generator sees the strongest context first, and to search over documentation, support tickets, internal knowledge bases or product catalogs. It also fits evaluation and data curation work, where you need a consistent relevance ordering across a large set of query and document pairs.
You call it through api14's OpenAI-compatible API using the model id "mongodb/rerank-2.5". Because the endpoint follows the OpenAI format, you can point an existing OpenAI client at api14 by changing the base URL and passing that model id in your request. The model lists no supported parameters, so there are no sampling or generation options to tune; you send your query and documents and receive the reranked ordering back. The 32,000 token context is the budget to keep in mind when deciding how many candidates to submit in a single call.
If your task is text generation, summarization, chat or tool use, a reranker is the wrong tool and a generative model will fit better. rerank-2.5 is best treated as one stage in a pipeline rather than a standalone assistant, and it is most valuable when you already have a candidate set that needs sharper ordering.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.post("/rerank", { body: {
"model": "mongodb/rerank-2.5",
"query": "What is the capital of France?",
"documents": [
"Paris is the capital of France.",
"Berlin is in Germany."
],
"top_n": 1
} });
console.log(result.results);
This community has no topics yet. Open the first one about VoyageAI by MongoDB: rerank-2.5.