Text Embedding 004

by Google

Google state-of-the-art 768-dimensional text embedding model with leading retrieval and semantic classification benchmark scores.

Parameters

768-dim

Context Length

About this model

Text Embedding 004 is Google's state-of-the-art text embedding model, generating 768-dimensional vector representations with industry-leading retrieval accuracy on MTEB benchmarks. Backs OpenAI-compatible /v1/embeddings on DOS.AI.

Capabilities

Semantic searchRAG retrievalClassificationClustering

Use Cases

RAG knowledge bases
Semantic search
Document deduplication
Content recommendations

Model Details

Provider: Google
Model ID: text-embedding-004
Parameters: 768-dim
Context Length: 2K tokens
Category: embedding

API Usage

Use the DOS API to integrate Text Embedding 004 into your applications. Our API is compatible with OpenAI's client libraries for easy migration.

Model ID

text-embedding-004

Python

python

from dos import DOS

client = DOS()

response = client.chat.completions.create(
    model="text-embedding-004",
    messages=[
        {"role": "user", "content": "Hello, how are you?"}
    ]
)

print(response.choices[0].message.content)

cURL

bash

curl https://api.dos.ai/v1/chat/completions \
  -H "Authorization: Bearer $DOS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "text-embedding-004",
    "messages": [
      {"role": "user", "content": "Hello, how are you?"}
    ]
  }'

Node.js

javascript

import DOS from 'dos-ai';

const client = new DOS();

const response = await client.chat.completions.create({
  model: "text-embedding-004",
  messages: [
    { role: "user", content: "Hello, how are you?" }
  ]
});

console.log(response.choices[0].message.content);

View full API reference

Related Models

DOS.AI

Qwen3 Embedding 4B

Self-hosted multilingual text embedding (2560 dims), strong on Vietnamese. Priced at our OpenRouter fallback cost.

$0.02 / 1M tokens

Google

Gemini Embedding 001

Google high-dimensional embedding model optimized for semantic search and Retrieval-Augmented Generation (RAG).

$0.02 / 1M tokens

Google

Multimodal Embedding 001

Google multimodal embedding model generating shared vector spaces for text, images, and video search.

$0.02 / 1M tokens