Skip to content

Developers

One integration. Every model in the catalog.

A consistent request format, one authentication method, and stable model IDs — so switching models is a one-line change.

Send a request

Pass a model ID and a list of messages. Change the model ID to route the same request to a different model.

Browse model IDs
import os
import requests

response = requests.post(
    "https://api.modelport.example/v1/chat/completions",
    headers={
        "Authorization": f"Bearer {os.environ['MODELPORT_API_KEY']}",
        "Content-Type": "application/json",
    },
    json={
        "model": "modelport-demo/atlas-chat",
        "messages": [
            {"role": "user", "content": "Summarize this support ticket in two sentences."}
        ],
        "max_tokens": 256,
    },
    timeout=30,
)
response.raise_for_status()
print(response.json()["choices"][0]["message"]["content"])
Illustrative request — endpoint pending configuration.

Sample response

response.json
{
  "id": "req_demo_01",
  "model": "modelport-demo/atlas-chat",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "The customer cannot reset their password..."
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "input_tokens": 182,
    "output_tokens": 41,
    "total_tokens": 223
  }
}

Sample error

error.json
{
  "error": {
    "type": "rate_limit_exceeded",
    "message": "Too many requests. Retry after the period in the Retry-After header.",
    "request_id": "req_demo_02"
  }
}