ai-agents-claude-thinking-conf-claude-3-7-sonnet-s-thinking-parameter-causes-unsu-dd731f64

THEN When using Claude 3.7 Sonnet with the thinking parameter, ensure budget_tokens is >= 1024 and temperature is set to 1. Use the model without the 'litellm_proxy/' prefix (e.g., model='claude-3-7-sonnet-20250219') or set litellm.drop_params=True in your proxy config to drop unsupported parameters.

response = litellm.completion(
    api_key="sk-...",
    base_url="http://localhost:4000",
    model='claude-3-7-sonnet-20250219',
    messages=[{"role": "user", "content": "Hello"}],
    max_tokens=2001,
    temperature=1,
    thinking={"type": "enabled", "budget_tokens": 2000}
)