API overview

All endpoints live under https://apirouter.pleum.ai/v1 and are compatible with the OpenAI API.

Base URL#

base url

https://apirouter.pleum.ai/v1

If you use the OpenAI SDK, put the address above in base_url (Python) or baseURL (JS).

Authentication#

Every inference request carries an API key in the Authorization header. Issue API keys under Dashboard → API Keys; they start with the plm_ prefix.

header

Authorization: Bearer plm_xxxxxxxxxxxxxxxx

An API key is shown only once at issuance, and only its hash is stored on the server. If a key leaks, disable it immediately from the dashboard. Per key you can restrict allowed models and a monthly budget cap.

Endpoints#

POST/v1/chat/completions

Creates a chat completion. Reference →

GET/v1/models

Lists available models and their pricing. Reference →

GET/v1/credits

Retrieves your credit balance. Reference →

GET/v1/fx-rate

Retrieves today's applied USD/KRW FX rate (Korea Eximbank base rate). No authentication required.

POST/v1/3d/generations

Submits an asynchronous 3D generation task to BytePlus ModelArk. Reference →

Error format#

Errors are returned in the detail field along with an HTTP status code. Structured errors (such as insufficient credits) include an error code string.

402 Payment Required

{
  "detail": {
    "error": "insufficient_credits",
    "message": "Insufficient credits: ₩120 < ₩540"
  }
}

For the full list of error codes, see the error codes doc.