OpenAI-compatible API

OpenAI-compatible LLM API

Most existing apps, SDKs and tools need no code change. Point base_url at the Tokenjoy gateway, use an API key issued by the gateway, and call any model in the catalog.

Request a demo

Three steps

Create an API key under “My keys” in the console and set its quota and allowed models
Set the SDK base_url to https://app.tokenjoy.me/v1
Put a catalog model name in the model parameter; switching models changes only that one value

Examples

                  curl https://app.tokenjoy.me/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer tj-..." \
  -d '{"model":"deepseek-v4-pro","messages":[{"role":"user","content":"Hello!"}]}'

                  from openai import OpenAI
client = OpenAI(api_key="tj-...", base_url="https://app.tokenjoy.me/v1")
resp = client.chat.completions.create(model="deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello!"}])
print(resp)

Capabilities depend on model and channel

Streaming, function calling and vision depend on the model and channel you pick. Besides chat, image, audio, video (async, two steps), embedding and rerank models are supported, and the example changes with the model type.

Keys and quota

An API key is shown once after creation, so copy it right away
Each key can have its own quota, expiry and model allowlist
After a forced rotation, the old key stops working after 24 hours

FAQ

Which SDKs work with Tokenjoy?

Any SDK or tool that lets you set the base_url of an OpenAI- or Anthropic-compatible API, such as the official OpenAI Python and Node SDKs.