Temperature

A setting that controls how much randomness goes into choosing each token — low is predictable, high is varied.

Generative AI Beginner 1 min read

Definition

At each step a model produces a probability distribution over the next token. Temperature reshapes that distribution before sampling. Below 1 it sharpens it, concentrating probability on the likeliest options; above 1 it flattens it, giving unlikely tokens more chance.

Temperature 0 is effectively deterministic — always take the most likely token — which is what you want for extraction, classification and anything you intend to parse. Higher values suit brainstorming and creative writing.

It does not control correctness. A low temperature makes a model consistently wrong rather than occasionally wrong.

How it relates to other terms

See also Hallucination When a model states something false with the same fluency and confidence it uses for things that are true. Part of Inference Running a trained model to get an output — as opposed to training, which is producing the model in the first p...
Referenced by
Hallucination See also Inference See also

Related terms

Diffusion Model

A generative model that learns to reverse a gradual noising process, turning random noise into an image step by step.

Advanced Generative AI

Prompt Engineering

Writing the input to a model deliberately — with context, examples and constraints — to get a more reliable output.

Beginner Generative AI

Was this definition helpful?

Definitions are reviewed by our editorial team. Spotted a problem? Tell us.

Keep reading

Browse the glossary Suggest a correction