BETA

Docs
Documentation

Python Embeddings

POST /v1/embeddings is live and proxied the same generic way as every other route.

Embedding-capable models come and go with the live catalog. Check GET /v1/models (or the Models page) for a current embedding model id before calling this endpoint — a model that isn't in the catalog returns 404 Model not found, same as every other /v1/* route.
main.py
from openai import OpenAI

client = OpenAI(base_url="https://apily.pro/v1", api_key="apily_your-api-key")

response = client.embeddings.create(
    input="Your text string goes here",
    model="<embedding-model-id-from-GET-/v1/models>",
)
print(response.data[0].embedding)

Response shape

A successful call follows the standard OpenAI convention: object: "list", data[].embedding (a float array) and usage.prompt_tokens.