API Reference

Embeddings

POSThttps://api.tokenbay.com/v1/embeddings

OpenAI-compatible embeddings and legacy compatibility path

OpenAI-compatible Path

openai

POST/v1/embeddings

Request

schema

modelstringRequired

Console-visible embedding model ID.

inputstring | string[] | arrayRequired

Input to embed.

dimensionsintegerOptional

Support depends on the actual upstream model.

Response

upstream-defined

Field semantics, returned vector dimensions, and features such as dimensions depend on the scheduled upstream model.

Legacy Compatibility Path

gemini

POST/v1/engines/{model}/embeddings

Request

schema

modelpath stringRequired

Model name in the URL path.

inputstring | string[]Required

Input to embed.

Response

upstream-defined

This path exists for clients that still use /v1/engines/{model}/embeddings.

OpenAI-compatible Path

bashOpenAI embeddings

Language

curl -i -X POST https://api.tokenbay.com/v1/embeddings \
  -H "Authorization: Bearer sk-XXXXXXX" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MODEL_ID_FROM_CONSOLE",
    "input": "hello"
  }'

Legacy Compatibility Path

bashLegacy compatibility path

Language

curl -i -X POST https://api.tokenbay.com/v1/engines/gemini-2.5-flash/embeddings \
  -H "Authorization: Bearer sk-XXXXXXX" \
  -H "Content-Type: application/json" \
  -d '{"input":"hello"}'

OpenAI-compatible Path

Request

Response

Legacy Compatibility Path

Request

Response

Related