Playground

Send a request to a deployed model through the regional inference endpoint.

Model

System prompt

Optional — sets the assistant's behavior.

User prompt

Temperature

Max tokens

Stream response

View as code

curl

curl /v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "x-api-key: ccsio_YOUR_API_KEY" \
  -d '{
  "model": "MODEL_ID",
  "messages": [
    {
      "role": "user",
      "content": ""
    }
  ],
  "stream": true,
  "temperature": 0.7,
  "max_tokens": 512
}'

Python (openai SDK)

from openai import OpenAI

client = OpenAI(
    base_url="https://<your-region>.ccsio.io/v1",
    api_key="ccsio_YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="MODEL_ID",
    messages=[
        {
            "role": "user",
            "content": ""
        }
    ],
    stream=True,
    temperature=0.7,
    max_tokens=512,
)

No requests yet

Send a prompt to see the model's reply stream in here. Nothing here is saved once you leave the page.