GPT-5.6 Sol

Previous-generation reasoning flagship.

A long-context rate applies above 272K input tokens, so the window exceeds 272K; the exact maximum and output limit are not published.

Make your first request

{
  "model": "gpt-5.6-sol",
  "messages": [
    {
      "role": "user",
      "content": "Hello, world."
    }
  ]
}