Inference API

By default, all apps on Cerebrium are accessible via authenticated POST requests. Authentication is handled using the JWT token available in the API Keys section of your dashboard.

The POST request follows the structure below, where {function} is the name of the function you want to invoke. For example, in this case, the function predict() from main.py is being called.

curl --location --request POST 'https://api.cortex.cerebrium.ai/v4/p-xxxxx/{app-name}/{function}' \
--header 'Authorization: Bearer <JWT_TOKEN>' \
--header 'Content-Type: application/json' \
--data '{
  "function_param": "data"
}'

Responses follow this format:

{
    "run_id": "52eda406-b81b-43f5-8deb-fcf80dfsb74b",
    "run_time_ms": 326.34
    "result": {<data>}
}

Successful requests return a status code of 200, while errors result in a status code of 500.

To return custom status codes such as 422, 404, etc., include the status_code field in the JSON response from your main.py.