Reasoning Model (System 2)
A Reasoning Model is an AI designed to 'think' before it speaks. Unlike standard chatbots that predict the next word instantly, these models generate a hidden 'chain of thought' to plan, verify, and correct their logic before outputting a final answer.
Why it Matters
This matters for solving complex math, coding, and scientific problems where accuracy is more important than speed.
How It Works
- 1
Often referred to as 'System 2' thinking (slow, deliberate) vs 'System 1' (fast, instinctive).
- 2
These models scale 'test-time compute,' meaning they spend more computational resources during inference to explore search trees and self-correct via reinforcement learning strategies.
Real-World Example
OpenAI o1 and DeepSeek R1 are reasoning models. If you ask 'How many Rs are in strawberry?', a standard model might guess wrong instantly, but a reasoning model will explicitly count the characters in its internal thought process before answering correctly.