Hadrian is experimental alpha software. Do not use in production.
Hadrian

Embeddings

Generate vector embeddings for text input. Use for semantic search, clustering, and similarity comparisons. OpenAI-compatible.

Create embeddings

POST
/api/v1/embeddings
AuthorizationBearer <token>

API key authentication using Bearer token format

In: header

Request Body

application/json

dimensions?integer|null

Number of dimensions for output embeddings

Formatint64
encoding_format?null|EncodingFormat
input*string||||

Input to embed

input_type?string|null

Hadrian Extension: Input type hint for embedding providers that support it (e.g., Cohere)

model*string

Model to use for embedding

provider?object

Hadrian Extension: Provider routing configuration

user?string|null

User identifier for abuse detection

Response Body

application/json

application/json

curl -X POST "https://loading/api/v1/embeddings" \  -H "Content-Type: application/json" \  -d '{    "dimensions": 1024,    "input": [      "First document to embed",      "Second document to embed",      "Third document to embed"    ],    "model": "openai/text-embedding-3-large"  }'
Empty
{
  "error": {
    "code": "routing_error",
    "message": "Model 'invalid-embedding-model' not found"
  }
}
{
  "error": {
    "code": "provider_error",
    "message": "Upstream provider returned error"
  }
}