siglip-so400m-patch14-384

Name: siglip-so400m-patch14-384
Author: google

by google

SigLIP SO400M: shape-optimized image-text encoder, a top-quality CLIP alternative

877Mparams

HuggingFace Run it on your own data, free

Identifiers

Model ID

google/siglip-so400m-patch14-384

Feature URI

Overview

SigLIP SO400M/14 at 384px pairs Google's sigmoid image-text loss with the SoViT-400M 'shape-optimized' backbone: a compute-efficient ViT that punches well above its 400M parameter count. It is one of the strongest open image-text encoders for zero-shot classification and retrieval, and a common default when teams want better accuracy than CLIP without going to billion-parameter models.

On Mixpeek, SigLIP SO400M is a visual embedding extractor for image and video-frame search, with fine-grained visual understanding that helps on detailed product, scene, and style queries.

Architecture

Shape-optimized ViT (SoViT-400M/14) image encoder at 384px with a paired text encoder, trained with the sigmoid (SigLIP) loss instead of softmax contrastive, which scales better and improves zero-shot accuracy at a given compute budget.

Mixpeek SDK Integration

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so siglip-so400m-patch14-384 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.