Interact with various LLMs models

{
  "model": "gpt-5-nano",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}

{
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Blablabla, shut the fuck up"
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 1,
    "completion_tokens": 10,
    "total_tokens": 11
  }
}