Top-P, also known as nucleus sampling, is a parameter that controls which tokens the model considers when generating each word. It sets a cumulative probability threshold — for example, Top-P of 0.9 means the model only considers the smallest set of tokens whose combined probability reaches 90%, filtering out unlikely options. It works alongside temperature to fine-tune output randomness.
With Top-P set to 0.1, the model only picks from the most likely next words, producing very predictable text. With Top-P at 0.95, the model considers a wider range of plausible words, allowing for more natural variation in phrasing.
Build polished, copy-ready prompts in under 60 seconds with SurePrompts.
Try SurePrompts