{
  "id": "reasoning-model",
  "code": "PTL-0083",
  "term": "Reasoning Model",
  "aliases": [
    "large reasoning model",
    "LRM",
    "thinking model"
  ],
  "category": "test-time",
  "definition": "A reasoning model is a language model trained, typically with reinforcement learning, to produce an extended internal chain of thought before answering, so that it improves with more thinking time on math, coding, and planning tasks.",
  "description": "OpenAI's o1, announced in 2024, popularized the category, and DeepSeek-R1 showed that reinforcement learning with verifiable rewards could produce long reasoning behaviors in an openly released model. Prompting such models differs from prompting standard models, because step-by-step instructions and few-shot reasoning examples are often unnecessary.",
  "example": null,
  "broader": [],
  "narrower": [
    "extended-thinking"
  ],
  "related": [
    "chain-of-thought",
    "test-time-compute-scaling",
    "self-taught-reasoner"
  ],
  "introduced": 2024,
  "sources": [
    {
      "title": "Learning to reason with LLMs",
      "authors": "OpenAI",
      "year": 2024,
      "url": "https://openai.com/index/learning-to-reason-with-llms/"
    },
    {
      "title": "DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning",
      "authors": "DeepSeek-AI",
      "year": 2025,
      "url": "https://arxiv.org/abs/2501.12948"
    }
  ],
  "url": "https://protologue.com/t/reasoning-model/",
  "citation": "Protologue. (2026). Reasoning Model. In Protologue: A Taxonomy of Prompting and LLM Techniques (v1.0.0, PTL-0083). https://protologue.com/t/reasoning-model/"
}