📚 Glossary
Reasoning model
In one line: An AI model that explicitly 'thinks' before answering by generating a long chain of thought. Better at math, code and logic — slower than chat models.
A reasoning model is an LLM that has been trained to do extensive chain-of-thought reasoning before producing a final answer. These models 'think out loud' for many tokens, then collapse the thinking into a clean final answer.
Reasoning models are slower (5-30 seconds per answer) but much more accurate on:
- Math problems
- Algorithmic / competitive-programming code
- Logic puzzles and multi-step deductions
- Scientific reasoning
Examples: DeepSeek R1, OpenAI's o1 / o3 series, Gemini 2.5 Pro 'Deep Think'. Use them for hard problems; use regular chat models for everyday questions.
See it in action — ask any AI about reasoning model on AskAI.free.
Try it free →