[Models](/models)/[Embeddings](/models?category=embeddings)/google/embeddinggemma-300m

HFText EmbeddingsGemma License

# embeddinggemma-300m

by google

Google's efficient multilingual text embedding -- 100+ languages in 300M parameters

1.5Mdl/month

308Mparams

[HuggingFace](https://huggingface.co/google/embeddinggemma-300m)[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

`google/embeddinggemma-300m`

## Deploy embeddinggemma-300m

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

EmbeddingGemma is Google's compact text embedding model based on the Gemma 3 architecture, designed for deployment on edge devices while maintaining strong multilingual performance. At 308M parameters, it runs in under 200MB of RAM when quantized and achieves sub-22ms inference on EdgeTPU.

Despite its small size, it ranks as the highest-performing open multilingual text embedding model under 500M parameters on MTEB. It supports Matryoshka representations (768, 512, 256 or 128 dimensions) for flexible memory/quality tradeoffs. On Mixpeek, it provides a lightweight embedding option for high-throughput text indexing where GPU resources are limited.

## Architecture

Gemma 3-based encoder, 308M parameters, 2,048-token context. Matryoshka output: 768 dimensions, truncatable to 512, 256 or 128. Text only. Succeeded in September 2026 by EmbeddingGemma 2, which adds images, video and audio.

## Mixpeek SDK Integration

IngestRetrieve

// No extractor parameter takes a Hugging Face model id (checked against
// GET /v1/discovery/extractors, which returns 13), so embeddinggemma-300m 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: \[
        {
          document\_id: "asset-00412",
          // The vector name has to match a vector index on the collection.
          vectors: { "text-embedding": yourVector },
          payload: { source\_key: "archive/2026/asset-00412" },
        },
      \],
    }),
  },
);

// Managed alternative, if this exact model is not the requirement:
// text\_extractor@v1 runs intfloat/multilingual-e5-large-instruct
// (1024-d) over a bucket, with no inference of your own.

## Capabilities

*   100+ language support
*   Matryoshka dimension reduction (768, 512, 256 or 128)
*   Sub-200MB quantized footprint
*   EdgeTPU and mobile-optimized
*   2048 token context

## Use Cases on Mixpeek

On-device semantic search in mobile applications

High-throughput text indexing at scale

Multilingual document retrieval

Lightweight embedding for resource-constrained deployments

## Benchmarks

Dataset

Metric

Score

Source

MTEB multilingual v2

Mean(Task)

61.15

Google, as reported in the EmbeddingGemma 2 model card comparison

MTEB code v1

Mean(Task), NDCG@10

68.76

Same comparison (EmbeddingGemma 2: 78.68)

## Performance

Input SizeUp to 2,048 tokens

Embedding Dim768 (Matryoshka: 512, 256, 128)

GPU LatencyInput dependent

GPU ThroughputBatch dependent

GPU MemoryModel dependent

Google states it runs in under 200 MB of RAM when quantized. It publishes no GPU latency or throughput figures and we have not measured them, so none are quoted here.

## Common Pipeline Companions

[

google/embeddinggemma-2

The multimodal successor: same text quality, plus code, images, video and audio

](/model/google/embeddinggemma-2)[

Qwen/Qwen3-Reranker-8B

Rerank lightweight embeddings for precision

](/model/Qwen/Qwen3-Reranker-8B)[

BAAI/bge-m3

Compare compact vs full-size embeddings

](/model/BAAI/bge-m3)

## Frequently Asked Questions

### Should I use EmbeddingGemma or EmbeddingGemma 2?

Use EmbeddingGemma 2 for anything new. It matches the first model on multilingual text (61.36 against 61.15 on MTEB), is much better at code (78.68 against 68.76), adds images, video and audio, has a 4x longer context and is Apache-2.0. Its text-only configuration is 270M parameters, about the size of the original.

### What embedding sizes does EmbeddingGemma support?

768 dimensions, truncatable to 512, 256 or 128 through Matryoshka representation learning. Re-normalize after truncating, and use the same size for queries and documents.

### How do I use EmbeddingGemma with Mixpeek?

Run the model on your side and upsert the vectors into a Mixpeek collection, as in the example on this page; a retriever then searches them. For managed text embeddings without running a model, text\_extractor@v1 uses multilingual-e5-large-instruct.

## Explore on Mixpeek

[

More Embeddings models

Compare alternatives in this category

](/models?category=embeddings)[

Best Self-Hosted Embedding Models

Hand-picked tools & platforms compared

](/curated-lists/best-self-hosted-embedding-models)[

Instruction-Tuned Embeddings

Deep-dive technical guide

](/guides/instruction-tuned-embeddings)[

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

Organizationgoogle

FeatureText Embeddings

Output1024-dim vector

Modalitiesdocument, audio

RetrieverText Similarity

Parameters308M

LicenseGemma License

Downloads/mo1.5M

[

### Research Paper

EmbeddingGemma

arxiv.org](https://huggingface.co/google/embeddinggemma-300m)

### Build a pipeline with embeddinggemma-300m

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

[

BAAI/bge-large-en-v1.5

Text Embeddings

](/model/BAAI/bge-large-en-v1.5)[

sentence-transformers/all-MiniLM-L6-v2

Text Embeddings

](/model/sentence-transformers/all-MiniLM-L6-v2)[

Qwen/Qwen3-Embedding-0.6B

Text Embeddings

](/model/Qwen/Qwen3-Embedding-0.6B)[

nomic-ai/nomic-embed-text-v2-moe

Text Embeddings

](/model/nomic-ai/nomic-embed-text-v2-moe)

### Related in Embeddings

[

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)