nomic-embed-vision-v1.5

Name: nomic-embed-vision-v1.5
Author: nomic-ai

by nomic-ai

Compact image embedding model aligned to Nomic's text embedding space

1.3Mdl/month

92.9Mparams

HuggingFace Run it on your own data, free

Identifiers

Model ID

nomic-ai/nomic-embed-vision-v1.5

Feature URI

mixpeek://image_extractor@v1/nomic_embed_vision_v15

Overview

Nomic Embed Vision v1.5 is a high-download image embedding model built to share an embedding space with Nomic Embed Text v1.5. That makes it useful for text-to-image retrieval without running a large vision-language model for every image.

On Mixpeek, it fits high-volume image and video-frame search where an agent needs fast first-stage recall. Use it to retrieve candidate frames, product images, screenshots, or visual evidence before a VLM inspects the short list.

Architecture

Vision embedding model with a locked text-embedding alignment strategy similar to LiT. The Hugging Face model card lists 92.9M parameters, Apache 2.0 licensing, and image-feature-extraction support through Transformers with custom code.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so nomic-embed-vision-v1.5 runs
// on your side and the output is upserted through POST
// /v1/namespaces/{namespace_id}/documents/upsert. On Enterprise the other
// path is to upload the weights instead: POST /v1/namespaces/{id}/models
// accepts the huggingface format and a custom plugin loads them.
const res = await fetch(
  "https://api.mixpeek.com/v1/namespaces/ns_your_namespace/documents/upsert",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer API_KEY",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      collection_id: "col_your_collection",
      documents: [
        {
          document_id: "asset-00412",
          // The vector name has to match a vector index on the collection.
          vectors: { "image-embedding": yourVector },
          payload: { source_key: "archive/2026/asset-00412" },
        },
      ],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// image_extractor@v1 runs google/siglip-base-patch16-224
// (768-d) over a bucket, with no inference of your own.