Dictionary
Temperature
The randomness knob on an LLM's output distribution.
1 min readupdated 2026-07-04
/ quick answer
Temperature scales the logits before sampling. 0 makes the model near-deterministic (best for extraction, code, JSON); 0.7-1.0 makes it creative (best for brainstorming, copy). The randomness knob on an LLM's output distribution.
The randomness knob on an LLM's output distribution. Temperature scales the logits before sampling. 0 makes the model near-deterministic (best for extraction, code, JSON); 0.7-1.0 makes it creative (best for brainstorming, copy). In practice: A JSON extractor runs at temperature 0.1; a hook generator runs at 0.9. This dictionary node is part of the Onexial knowledge graph and links to related concepts, workflows and tools below.
Definition
Temperature scales the logits before sampling. 0 makes the model near-deterministic (best for extraction, code, JSON); 0.7-1.0 makes it creative (best for brainstorming, copy).
Example
A JSON extractor runs at temperature 0.1; a hook generator runs at 0.9.
Related Prompts
/ frequently asked
What is Temperature?
Temperature scales the logits before sampling. 0 makes the model near-deterministic (best for extraction, code, JSON); 0.7-1.0 makes it creative (best for brainstorming, copy).
What is an example of Temperature?
A JSON extractor runs at temperature 0.1; a hook generator runs at 0.9.
Why does Temperature matter for AI and automation?
The randomness knob on an LLM's output distribution. It connects to the workflows, prompts and tool stacks linked on this page, so you can move from definition to execution without leaving Onexial.
↳ connected nodes
Prompt↳ linked
Viral Hook Generator Prompt
Produce 10 scroll-stopping hooks for a topic and platform.
Prompt↳ linked
Grounded Answer Prompt
Force the model to answer only from provided sources, with citations.
Dictionary↳ linked
Few-Shot Prompting
Showing the model 2–5 examples of the task inside the prompt so it mirrors the pattern.
Dictionary↳ linked
Chain-of-Thought Prompting
Instructing a model to think step by step before answering.
Dictionary↳ linked
Zero-Shot Prompting
Asking a model to do a task with no examples in the prompt.
Dictionary↳ linked
Prompt Chaining
Pipelining LLM calls where each step's output feeds the next.