PromptForge
AI Glossary

Models & architecture

Reasoning model

Also known as: reasoning models, thinking model

Definition

A reasoning model is trained to generate extended internal deliberation before answering. It trades latency and cost for substantially better performance on maths, logic, coding and multi-step problems.

Reasoning models internalise what chain-of-thought prompting does externally. Rather than being told to think step by step, they are trained to produce long reasoning traces — often hidden from the user — before committing to an answer.

The gains on verifiable tasks are large: competition maths, algorithmic coding, scientific problems. The costs are also real. Responses take longer, consume many more tokens (you pay for reasoning tokens even when they are not shown), and the extra deliberation adds nothing on simple tasks.

Practical routing: use reasoning models for hard analytical problems, standard models for writing, summarisation, extraction and conversation.

Related terms

Put this into practice

Understanding the term is step one. Our free courses and tools let you actually use it.