Skip to main content
Kimi K3 always reasons before answering. How much it thinks is controlled by the top-level reasoning_effort parameter, which accepts three levels - “low”, “high” and “max” - and defaults to “max” when omitted.
reasoning_effort.py

Run the Example

1

Set up your virtual environment

2

Install dependencies

3

Export your Moonshot API key

4

Run the example

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