POST/v1/embeddings
Embeddings
POST /v1/embeddings: turn text into vectors with an embedding model, in the OpenAI format.
La documentazione è in inglese.
Creates embedding vectors for one or more texts. Works with the models in the embedding category; they list only /v1/embeddings in their endpoints.
Base URL: https://api.nexrouting.com/v1 (also https://nexrouting.com/api/v1).
Headers
AuthorizationstringrequiredAuthorization: Bearer <key>(orx-api-key: <key>).Content-Typestringapplication/json. The body is read as JSON either way.
Body
modelstringrequired- Model ID, such as
openai/text-embedding-3-small. inputstring | arrayrequired- A text, or an array of texts. One request with many texts counts once against your rate limit.
dimensionsinteger- Length of the vectors, for models that support it.
encoding_formatstringfloatorbase64, for models that support it.
There is no streaming on this endpoint. You pay for the input tokens at the model's input price.
Example request
curl https://api.nexrouting.com/v1/embeddings \ -H "Authorization: Bearer $NEXROUTING_API_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/text-embedding-3-small", "input": ["first text", "second text"] }'Example response
{ "object": "list", "data": [ { "object": "embedding", "index": 0, "embedding": [0.0123, -0.0456, 0.0789] }, { "object": "embedding", "index": 1, "embedding": [0.0234, -0.0567, 0.0891] } ], "model": "openai/text-embedding-3-small", "usage": { "prompt_tokens": 6, "total_tokens": 6 }}Vectors are shortened here; real ones have hundreds or thousands of numbers.
Errors
OpenAI error shape. A chat model on this endpoint returns 400 unsupported_endpoint. All codes are in Errors.