[Models](/models)/[Embeddings](/models?category=embeddings)/jinaai/jina-embeddings-v5-omni-nano-retrieval

HFVisual Embeddingscc-by-nc-4.0

# jina-embeddings-v5-omni-nano-retrieval

by jinaai

Compact omni retrieval embeddings for low-latency agent memory

3Kdl/month

8likes

948Mparams

[HuggingFace](https://huggingface.co/jinaai/jina-embeddings-v5-omni-nano-retrieval)[Run it on your own data, free](https://studio.mixpeek.com/namespaces/create?mode=managed&utm_source=homepage&utm_medium=model_page&utm_campaign=mvs_onramp)

Identifiers

Model ID

`jinaai/jina-embeddings-v5-omni-nano-retrieval`

## Deploy jina-embeddings-v5-omni-nano-retrieval

Single-tenant

Mixpeek has no managed extractor for this model. On a single-tenant deployment you upload the weights and a custom plugin serves them next to the rest of your pipeline.

[Read the customer-hosted guide](https://mixpeek.com/docs/customer-hosted/overview) [Talk to us about single-tenant](/demo)

## Overview

Jina Embeddings v5 Omni Nano Retrieval is the compact retrieval checkpoint in Jina's v5 omni family. It is designed for multimodal retrieval across text, images, audio, and video while keeping model size and serving cost low.

On Mixpeek, the nano retrieval model is useful when an agent needs broad first-stage recall over mixed media but has tight latency, budget, or edge deployment constraints. Use it as a fast candidate generator before a larger reranker or vision-language model inspects the final evidence.

## Architecture

Compact multimodal embedding model exposed through the Transformers feature-extraction pipeline. The retrieval checkpoint is tuned for query-document similarity across multiple input modalities.

## Mixpeek SDK Integration

IngestRetrieve

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so jina-embeddings-v5-omni-nano-retrieval 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.

## Capabilities

*   Text, image, audio, and video retrieval embedding support
*   Compact serving footprint for high-throughput pipelines
*   First-stage recall for agent memory and mixed-media search
*   Pairs naturally with multimodal rerankers

## Use Cases on Mixpeek

Low-latency retrieval over support clips, screenshots, and transcripts

Agent memory indexes where each tool call must stay within a budget

Mobile or edge-adjacent visual search with compact embeddings

Candidate generation before Qwen3-VL or Jina multimodal reranking

## Common Pipeline Companions

[

Qwen/Qwen3-VL-Reranker-2B

Rerank compact omni retrieval candidates

](/model/Qwen/Qwen3-VL-Reranker-2B)[

microsoft/Florence-2-large

Generate captions for retrieved screenshots

](/model/microsoft/Florence-2-large)

## Explore on Mixpeek

[

More Embeddings models

Compare alternatives in this category

](/models?category=embeddings)[

Best Multimodal Embedding Models

Hand-picked tools & platforms compared

](/curated-lists/best-multimodal-embedding-models)[

How CLIP, SigLIP & CLAP Work

Deep-dive technical guide

](/guides/contrastive-learning-how-clip-siglip-and-clap-work)[

Feature Extractors

See how Mixpeek runs models as extractors

](/extractors)[

Mixpeek Vector Store

Store & search embeddings at scale

](/mvs)[

Pricing

Usage-based pricing for pipelines

](/pricing)[

All Curated Lists

Compare models, APIs & infrastructure

](/curated-lists)

### Specification

FrameworkHF

Organizationjinaai

FeatureVisual Embeddings

Output768-dim vector

Modalitiesvideo, image

RetrieverVector Search

Parameters948M

Licensecc-by-nc-4.0

Downloads/mo3K

Likes8

[

### Research Paper

Jina Embeddings v5 Omni Nano Retrieval

arxiv.org](https://huggingface.co/jinaai/jina-embeddings-v5-omni-nano-retrieval)

### Build a pipeline with jina-embeddings-v5-omni-nano-retrieval

Add this model to a processing pipeline alongside other extractors. Combine with retrieval stages for end-to-end search.

[Run it on your own data, free](https://studio.mixpeek.com/namespaces/create?mode=managed&utm_source=homepage&utm_medium=model_page&utm_campaign=mvs_onramp)

### Alternative Models

[

openai/clip-vit-large-patch14

Visual Embeddings

](/model/openai/clip-vit-large-patch14)[

google/siglip-base-patch16-224

Visual Embeddings

](/model/google/siglip-base-patch16-224)[

google/siglip2-giant-opt-patch16-384

Visual Embeddings

](/model/google/siglip2-giant-opt-patch16-384)[

facebook/dinov2-large

Visual Embeddings

](/model/facebook/dinov2-large)

### Related in Embeddings

[

laion/clap-htsat-fused

Audio Embeddings

](/model/laion/clap-htsat-fused)[

facebook/pe-av-large

Audio Embeddings

](/model/facebook/pe-av-large)[

tsinghua-ee/WAVE-7B

Audio Embeddings

](/model/tsinghua-ee/WAVE-7B)[

facebook/encodec\_24khz

Audio Embeddings

](/model/facebook/encodec_24khz)