Temperature
Temperature is a setting between 0 and roughly 2 that controls how much randomness a model uses when choosing each next token.
Low temperature makes the model pick the most likely continuation almost every time, which suits extraction, classification, and code. Higher temperature widens the choice, which suits brainstorming and creative writing. Temperature does not make a model more or less accurate; it makes it more or less varied.
Related terms
- Top-p (nucleus sampling)
- Top-p is a sampling setting that restricts the model to the smallest set of next tokens whose combined probability exceeds a threshold, discarding the rest.
- Determinism
- A process is deterministic when the same input always produces exactly the same output, with no randomness involved.
More on core concepts
See the full glossary, read the guides, or put it into practice in the prompt builder.