What is Temperature?
Temperature is a setting that controls how random a model's word choices are — low for consistent, factual output; higher for variety and creativity.
At each step the model has a probability distribution over possible next tokens. Temperature reshapes it: near 0 the model almost always picks the most likely token (deterministic, repetitive, safe); around 0.7–1.0 it samples more freely (varied, occasionally surprising); above that it gets erratic.
It's an API/developer setting — chat interfaces usually hide it. If you're building something, use low temperature for extraction, classification and code, and moderate temperature for brainstorming and copy variants. It does not make the model "smarter" or more accurate; it only changes how it picks among options.
- 0–0.3 for anything that must be consistent (data extraction, JSON, code).
- 0.7–1.0 for ideation and multiple variants.
- If you want variety in a chat UI, ask for "five different angles" instead.
Use it right now
Ask our brain anything on the homepage — it remembers the whole conversation — or write a brief in the Studio and see the prompt it compiles to.