GET /models

List the models this key may call, or fetch one model's supported parameters and billing mode.

On this page

text

GET https://api.apione.ai/v1/models
GET https://api.apione.ai/v1/models/{model}/params

Listing models

Returns every model the current key may call. The list is already filtered by the key's allowed models; there are no query parameters.

bash

curl https://api.apione.ai/v1/models \
  -H "Authorization: Bearer $APIONE_API_KEY"

json

{
  "object": "list",
  "data": [
    { "id": "claude-sonnet-4-6", "object": "model", "created": 0, "owned_by": "maxx" },
    { "id": "gpt-5.6-sol", "object": "model", "created": 0, "owned_by": "maxx" }
  ]
}

id is the value for the request's model field. Prices, context length and the rest live in the model catalog.

Model parameters

Returns one model's billing mode and the extra parameters it accepts — mostly useful for video and image models billed per call or per second.

bash

curl https://api.apione.ai/v1/models/kling-2-5-turbo/params \
  -H "Authorization: Bearer $APIONE_API_KEY"

json

{
  "model": "kling-2-5-turbo",
  "billingMode": "duration",
  "chargeModes": ["usage", "request", "second", "image"],
  "params": ["resolution", "generation_mode", "quality"],
  "request": {
    "resolution": { "type": "string", "enum": ["720p", "1080p", "2k", "4k"] },
    "generationMode": { "type": "string", "enum": ["text_to_video", "image_to_video", "video_reference"] },
    "mode": { "type": "string", "enum": ["std", "pro", "2k", "4k"] }
  }
}

billingModestring

How the model bills: token by usage, duration by length, request per call, hybrid mixed; an empty string means the gateway decides.

chargeModesstring[]

Charge modes allowed for this model: usage / request / second / image.

paramsstring[]

Parameters the model declares; an empty array when it has none.

requestobject

Fields you may send in the request body and their allowed values; {} when unconstrained.

If the model doesn't exist or this key may not call it, the response is 404 with {"error": "model not found"}.

Tip

Fetch the model list once at startup to check the model IDs in your config are still available — far earlier than hitting 400 model_not_supported in production.

Contact us

Something unclear, or need a human? Pick the channel that fits your question.

GET /models

Listing models

Model parameters

Contact us

Email

WhatsApp

WeCom

Telegram

Support tickets