Astra Compute/Playground

Playground

Test prompts against your machine endpoints or the managed fallback network.

Request
0.70
Response

Run a prompt to see the response, latency, and cost here.

request payload
{
  "target": "machine",
  "machineId": "",
  "model": "llama-3.1-8b",
  "prompt": "Explain how GPU inference works in one paragraph.",
  "temperature": 0.7,
  "maxTokens": 512
}