Skip to main content
Toggle thinking mode with the use_thinking flag. use_thinking=True makes the model emit reasoning_content before its answer; use_thinking=False turns it off for a faster, cheaper response. Leaving it unset (None) uses the model default.
thinking_mode.py

Run the Example

1

Set up your virtual environment

2

Install dependencies

3

Export your Xiaomi MiMo API key

4

Run the example

Save the code above as thinking_mode.py, then run:
Full source: cookbook/90_models/xiaomi/thinking_mode.py