D4·Fix the code

Thinking configuration after an upgrade

This request errors on a current model. It was written when a fixed thinking budget was the way to control reasoning depth.

response = client.messages.create(
    model="claude-opus-5",
    max_tokens=16000,
    thinking={"type": "enabled", "budget_tokens": 8000},
    effort="high",
    temperature=0.2,
    messages=messages,
)

1.Which parameters are rejected on current frontier models?

2.Where does `effort` belong?

← Forcing JSON the old way Batch results that silently mismatch →