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/voyage-4-largeRuns on your balance at the prices above. New accounts start with free usage.
Settings
VoyageAI by MongoDB: voyage-4-large is a general-purpose and multilingual embedding model from VoyageAI by MongoDB. It sits in the embeddings category, so its job is to turn text into vector representations rather than to generate chat responses. It is built for retrieval quality, which makes it a natural fit for search and retrieval systems where the quality of the vector representation directly affects what gets retrieved.
The model accepts text as input and returns embeddings as output. It has a context length of 32,000 tokens and a maximum output of 28,800. Its design is enabled by Matryoshka learning and quantization-aware training, and it supports embeddings. Because it is multilingual and general-purpose, it can be applied to text in more than one language without switching to a separate model.
Practical use cases follow from those properties. Retrieval is the primary one: indexing documents, passages or records and embedding queries so that a search layer can rank results by semantic similarity. Multilingual corpora are a good fit, since the same model can embed content and queries across languages. Matryoshka learning means the embeddings can be used at different truncation levels, which is useful when you need to trade storage or latency against retrieval quality in a vector index. Quantization-aware training matters when embeddings are stored in compressed form, since it keeps quality closer to full precision vectors.
You call the model through api14's OpenAI-compatible API using the model id "mongodb/voyage-4-large". Because the endpoint follows the OpenAI format, you can point an existing OpenAI client at api14 and send embedding requests in the usual way, with the model id in the request body. The model supports no additional request parameters, so requests stay simple: the model id and the input text are what matter. The page's live price table shows the current rates.
One limit to keep in mind is that this is an embeddings model, not a text generator. If you need chat completions, reasoning, tool calling or any generated text output, a different model is the right choice. Voyage-4-large is for producing vectors, and its 32,000 token context bounds how much text can go into a single embedding request.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.embeddings.create({
"model": "mongodb/voyage-4-large",
"input": [
"The quick brown fox",
"Lorem ipsum"
]
});
console.log(result.data.map((d) => d.embedding.length));
This community has no topics yet. Open the first one about VoyageAI by MongoDB: voyage-4-large.