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-multimodal-3.5Runs on your balance at the prices above. New accounts start with free usage.
Settings
VoyageAI by MongoDB: voyage-multimodal-3.5 is a multimodal embedding model from VoyageAI by MongoDB. It belongs to the embeddings category, so its job is to turn content into vectors rather than to generate chat responses. It is aimed at developers who need a single embedding model that covers more than plain text.
The model accepts text and images as input and produces embeddings as output. It can vectorize text, images, and video individually, and it also handles content that interleaves all three modalities. The context window is 32,000 tokens, and the maximum output is 28,800. Because the output is always a vector representation, the model does not return prose, answers, or chat completions; it returns embeddings that you store or compare.
Those capabilities point to a few practical uses. You can embed text documents and images into one shared vector space for search and retrieval, which suits knowledge bases and catalogs that mix written material with visual assets. Because the model handles interleaved text, image, and video content, it also fits pages or records where the modalities appear together, such as product listings with descriptions and photos. The embeddings can support similarity matching, clustering, deduplication, and recommendation features built on top of your own vector store.
You call the model through api14's OpenAI-compatible API using the model id "mongodb/voyage-multimodal-3.5". Because the endpoint follows the OpenAI format, you can use existing OpenAI client libraries and point them at api14, sending text and image inputs and reading back the embedding vectors. The model lists no supported parameters, so plan on the defaults for this model and keep request configuration minimal; check the endpoint documentation for the exact request shape before wiring it into a pipeline.
Keep the embedding role in mind when choosing between models. This one does not generate text or carry on conversations, so tasks that need generated answers or reasoning call for a chat or completion model instead. It is the better fit when the requirement is to represent text, images, and mixed content as vectors for search and retrieval.
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-multimodal-3.5",
"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-multimodal-3.5.