Models & inference

Temperature

A 0-to-1 dial that controls how random the model's output is. Lower = more deterministic.

also known as: sampling temperature

In depth

Temperature 0.0 means the model picks the most likely next token every time, which is what you want for fact-checking, classification, and any task where you want repeatable answers. Temperature 0.7-1.0 produces more variety, useful for creative writing and brainstorming. Most agent setups run between 0.0 and 0.3.

Related concepts

LLMA neural network trained on text that takes a prompt and returns text, optionally including structured tool calls.Tool useThe LLM capability to emit a structured call to an external function, which the runtime executes and feeds back as context.

Newsletter

Get the next post in your inbox.

Engineering notes from the Digitorn team. No marketing, no launch announcements, no "10 prompts that will change your life". Just the things we write that we'd want to read.

More in Models & inference

Context window/glossary/context-window Frontier model/glossary/frontier-model Inference/glossary/inference LLM/glossary/llm Open-weight model/glossary/open-weight-model Streaming/glossary/streaming