What are "reasoning" AI models?
“Reasoning” AI models1 are LLMs that spend some time and compute tokens “thinking” before answering queries. Examples include OpenAI’s o1 and o3, DeepSeek’s R1, Anthropic’s Claude 3.7 and Google DeepMind’s Gemini Flash Thinking.
These models use a process similar to chain-of-thought prompting to reflect on their own output. This allows these models to perform better on certain tasks such as analytical reasoning.
Caption: DeepSeek’s R1 reflecting on the query it is asked
Reasoning models require substantially more compute than non-reasoning models, which increases both their cost per token and their environmental impact.
As of Q1 2025, these models are rather new and there is no standardized name for this type of model. They have been called simulated reasoning models, large reasoning models and enhanced reasoning models. ↩︎