Create embeddings · DeutschlandGPT

Creates an embedding vector representing input text. Compatible with the OpenAI /v1/embeddings API.

Pass a single string or an array of strings. The response contains one embedding vector per input.

POSThttps://api.deutschlandgpt.de/v2/embeddingsTry it

Example request

curl https://api.deutschlandgpt.de/v2/embeddings \
  -X POST \
  -H "Authorization: Bearer $DGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": "The quick brown fox jumps over the lazy dog.",
  "model": "text-embedding-3-large",
  "encoding_format": "float",
  "dimensions": 0,
  "task_type": "SEMANTIC_SIMILARITY"
}'

curl https://api.deutschlandgpt.de/v2/embeddings \
  -X POST \
  -H "Authorization: Bearer $DGPT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": "The quick brown fox jumps over the lazy dog.",
  "model": "text-embedding-3-large",
  "encoding_format": "float",
  "dimensions": 0,
  "task_type": "SEMANTIC_SIMILARITY"
}'

import os, requests

response = requests.post(
    "https://api.deutschlandgpt.de/v2/embeddings",
    headers={"Authorization": f"Bearer {os.environ['DGPT_API_KEY']}"},
    json={
        "input": "The quick brown fox jumps over the lazy dog.",
        "model": "text-embedding-3-large",
        "encoding_format": "float",
        "dimensions": 0,
        "task_type": "SEMANTIC_SIMILARITY"
    },
)

print(response.json())

import os, requests

response = requests.post(
    "https://api.deutschlandgpt.de/v2/embeddings",
    headers={"Authorization": f"Bearer {os.environ['DGPT_API_KEY']}"},
    json={
        "input": "The quick brown fox jumps over the lazy dog.",
        "model": "text-embedding-3-large",
        "encoding_format": "float",
        "dimensions": 0,
        "task_type": "SEMANTIC_SIMILARITY"
    },
)

print(response.json())

const response = await fetch('https://api.deutschlandgpt.de/v2/embeddings', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.DGPT_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "input": "The quick brown fox jumps over the lazy dog.",
    "model": "text-embedding-3-large",
    "encoding_format": "float",
    "dimensions": 0,
    "task_type": "SEMANTIC_SIMILARITY"
  }),
});

console.log(await response.json());

const response = await fetch('https://api.deutschlandgpt.de/v2/embeddings', {
  method: 'POST',
  headers: {
    Authorization: `Bearer ${process.env.DGPT_API_KEY}`,
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    "input": "The quick brown fox jumps over the lazy dog.",
    "model": "text-embedding-3-large",
    "encoding_format": "float",
    "dimensions": 0,
    "task_type": "SEMANTIC_SIMILARITY"
  }),
});

console.log(await response.json());

Request body

inputstring | string[]required

Text(s) to embed. Each string must be within the model's max token limit.

modelstringoptional

Embedding model ID.

encoding_formatstringoptional

float returns a number array; base64 returns a base64-encoded Float32Array binary.

dimensionsintegeroptional

Desired output dimensions. Only supported by select models (e.g. text-embedding-3-large, gemini-embedding-001). Must be within the model's supported range.

task_typestringoptional

Task type hint to optimize embeddings for. Only supported by Gemini embedding models.

Response

200application/json

{
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        null
      ],
      "index": 0
    }
  ],
  "model": "string",
  "usage": {
    "prompt_tokens": 0,
    "total_tokens": 0
  }
}

{
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        null
      ],
      "index": 0
    }
  ],
  "model": "string",
  "usage": {
    "prompt_tokens": 0,
    "total_tokens": 0
  }
}

Embedding response

objectstring

dataobject[]

objectstring

embeddingnumber[] | string

indexinteger

Index of the input string this embedding corresponds to

modelstring

usageobject

prompt_tokensinteger

total_tokensinteger

Response codes

200

Embedding response

object

400

Unsupported dimensions or task_type

Error

401

Unauthorized

402

Insufficient credits

404

Model not found