道练习

Tuning reasoning effort

By default, V4-Pro thinks at "high" effort. If you need it to push harder on a particularly tough problem, set reasoning_effort="max" and the model will spend longer working through the steps.

In this exercise, you'll send the same math_problem from the previous exercise — this time with reasoning_effort="max" — and compare how the response is laid out.

说明

100 XP

Create a request to send math_problem to "deepseek-ai/DeepSeek-V4-Pro".
Set reasoning_effort to "max" to crank up the thinking intensity.

.css-6su6fj{-webkit-flex-shrink:0;-ms-flex-negative:0;flex-shrink:0;}道练习

说明

道练习