Skip to content
Last Dictionary
Plane
Language Models & TransformersTraining & Post-trainingReinforcement Learning
emerging

reasoning model

nounAlso: thinking model, large reasoning model, LRM, reasoning LLM, reasoner

In plain wordsAn AI model trained to work through a problem step by step in hidden 'thinking' text before it answers. Letting it think longer generally makes it better at math, coding and planning.

  1. A large language modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. trained, chiefly by reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time. on tasks with checkable answers (reinforcement learning with verifiable rewardsreinforcement learning with verifiable rewards, n.: Training an AI by letting it attempt problems whose answers can be checked automatically — a math result, a passing program — and rewarding it only when the check succeeds.), to generate a long reasoning tracereasoning trace, n.: The written-out 'thinking' an AI reasoning model produces before its final answer — notes to itself full of attempts, checks and corrections, which users may or may not get to see. — decomposing the problem, trying approaches, checking intermediate results and backtracking — before producing its final answer. Unlike prompting an ordinary model to show its work (chain of thoughtchain of thought, n.: Getting an AI to write out its reasoning step by step before giving an answer, the way a student shows their working. This often makes answers to math and logic problems more accurate.), the behavior is learned and spontaneous, and accuracy rises with the number of reasoning tokens spent, so better answers can be bought per question with extra computation (test-time computetest-time compute, n.: The computing effort an AI model spends while answering a question, as opposed to while being trained. Letting a model think longer or try many answers can make it markedly smarter.) instead of only by building and training a bigger model. Developers can usually set a thinking budgetthinking budget, n.: A limit, set by whoever uses an AI model, on how much private step-by-step reasoning it may do before answering: more thinking usually means better answers but higher cost and slower replies. or effort level. Reasoning models made large gains on competition mathematics, programming and science benchmarks, but cost more per query, respond more slowly and may overthinkoverthinking, n.: When an AI reasoning model spends far more thinking than a problem needs — pages of deliberation for a simple sum — wasting time and money and sometimes talking itself out of a right answer. easy questions. Whether their traces reflect genuine reasoning, and how faithfully they reveal what drives the answer, is debated; see chain-of-thought faithfulnesschain-of-thought faithfulness, n.: Whether the reasoning an AI writes out really reflects how it reached its answer, rather than being an after-the-fact story that leaves out what actually drove the decision..

    The reasoning model thought for ninety seconds before it found the off-by-one error.

Origin

OpenAI's o1-preview (12 September 2024) introduced the category, reporting that performance improved with both reinforcement-learning compute and time spent thinking; days later Karthik Valmeekam, Kaya Stechly and Subbarao Kambhampati called o1 a new kind of model, a large reasoning model (LRM). DeepSeek-R1DeepSeek-R1, prop. n.: A free, downloadable AI reasoning model released by the Chinese company DeepSeek in January 2025. It matched leading American models at math and coding for a fraction of the reported cost, shaking the industry. (January 2025) published an open-weight recipe with visible traces, and Anthropic's Claude 3.7 Sonnet (February 2025) and Google's Gemini 2.5 Flash (April 2025) followed as hybrid modelshybrid reasoning model, n.: A single AI model that can either answer right away or first think at length, with the user or developer choosing how much thinking to allow..

Usage note

All LLMs reason to some degree; the term marks models trained to deliberate at length before answering. Apple researchers' 'The Illusion of Thinking' (2025) argued that accuracy collapses beyond a certain puzzle complexity; critics disputed its methods. What such models are doing remains debated.

Related

See also
  • chain of thoughtchain of thought, n.: Getting an AI to write out its reasoning step by step before giving an answer, the way a student shows their working. This often makes answers to math and logic problems more accurate.
  • test-time computetest-time compute, n.: The computing effort an AI model spends while answering a question, as opposed to while being trained. Letting a model think longer or try many answers can make it markedly smarter.
  • reasoning tracereasoning trace, n.: The written-out 'thinking' an AI reasoning model produces before its final answer — notes to itself full of attempts, checks and corrections, which users may or may not get to see.
  • extended thinkingextended thinking, n.: A setting in some AI models, notably Anthropic's Claude, that lets the model write out private step-by-step reasoning before its final answer, improving results on hard problems.
  • thinking budgetthinking budget, n.: A limit, set by whoever uses an AI model, on how much private step-by-step reasoning it may do before answering: more thinking usually means better answers but higher cost and slower replies.
  • reinforcement learning with verifiable rewardsreinforcement learning with verifiable rewards, n.: Training an AI by letting it attempt problems whose answers can be checked automatically — a math result, a passing program — and rewarding it only when the check succeeds.
  • DeepSeek-R1DeepSeek-R1, prop. n.: A free, downloadable AI reasoning model released by the Chinese company DeepSeek in January 2025. It matched leading American models at math and coding for a fraction of the reported cost, shaking the industry.
  • System 1 and System 2System 1 and System 2, phr.: A popular way of describing two modes of thinking: fast, automatic and intuitive (System 1) versus slow, effortful and deliberate (System 2). AI researchers borrow it to contrast instant answers with step-by-step reasoning.
Contrast with
  • chat modelchat model, n.: A language model trained to hold a back-and-forth conversation, keeping track of who said what and replying as a helpful assistant. It is the engine inside chat apps.

Sources

  1. OpenAI (2024). Learning to Reason with LLMs. OpenAI blog, 12 September 2024. openai.com/index/learning-to-reason-with-llms
  2. DeepSeek-AI (2025). DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning. arXiv:2501.12948.
  3. Valmeekam, K., Stechly, K. & Kambhampati, S. (2024). LLMs Still Can't Plan; Can LRMs? A Preliminary Evaluation of OpenAI's o1 on PlanBench. arXiv:2409.13373. arxiv.org/abs/2409.13373

Mentioned in

ARC-AGIARC-AGI, prop. n.: Puzzles made of colored grids — and, since 2026, small video-game-like worlds — that people solve easily but AI long found hard, designed to test whether a machine can learn a new skill from just a few examples., chain of thoughtchain of thought, n.: Getting an AI to write out its reasoning step by step before giving an answer, the way a student shows their working. This often makes answers to math and logic problems more accurate., ChatGPTChatGPT, prop. n.: OpenAI's AI chatbot, released free to the public in November 2022, which answers questions, writes and converses in everyday language. Its sudden popularity began the current AI boom., deep researchdeep research, n.: An AI feature that spends minutes rather than seconds on a question — planning, running dozens of web searches, reading many sources — and then writes a long report with citations., DeepSeekDeepSeek, prop. n.: A Chinese artificial intelligence company whose free, cheaply trained models startled the world in January 2025 by nearly matching the best American systems., GeminiGemini, prop. n.: Google's family of AI models and its chatbot assistant, launched in December 2023. Gemini models handle text, images, audio and video and power many Google products., GPT-5GPT-5, prop. n.: OpenAI's flagship AI model from August 2025, which decides by itself whether to answer quickly or think longer. It replaced earlier models in ChatGPT and was followed by frequent upgrades., inference costinference cost, n.: What it costs to use an AI model after it has been built: the computing power, electricity and hardware needed every time it answers, usually billed per chunk of text., large language modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code., OpenAIOpenAI, prop. n.: The American company that makes ChatGPT and the GPT models. It began in 2015 as a nonprofit research lab and is now a for-profit company overseen by a nonprofit foundation., reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time., AIMEAIME, prop. n.: A hard US high-school math contest whose 15 questions each have a whole-number answer; AI labs use each year's new papers to test their models' mathematical reasoning., ByteDanceByteDance, prop. n.: The Chinese company that owns TikTok, which also runs Doubao, China's most popular AI chatbot, and builds image and video generators., chain-of-thought monitoringchain-of-thought monitoring, n.: Having a program — often another AI — read the step-by-step reasoning an AI writes before it acts, to catch signs that it intends to cheat, deceive or cause harm., chat modelchat model, n.: A language model trained to hold a back-and-forth conversation, keeping track of who said what and replying as a helpful assistant. It is the engine inside chat apps., CodexCodex, prop. n.: A name OpenAI has used twice: first for a 2021 AI model that wrote code and powered the original GitHub Copilot, then from 2025 for its family of coding agents., Daniel KahnemanDaniel Kahneman, prop. n.: An Israeli-American psychologist who showed how human judgment systematically errs and popularized the idea of fast, intuitive versus slow, deliberate thinking, now used to describe AI reasoning., DeepSeek momentDeepSeek moment, n.: The shock of January 2025, when a little-known Chinese company released a free AI model nearly as good as the best American ones at a fraction of the reported cost, briefly wiping hundreds of billions off chip stocks., DeepSeek-R1DeepSeek-R1, prop. n.: A free, downloadable AI reasoning model released by the Chinese company DeepSeek in January 2025. It matched leading American models at math and coding for a fraction of the reported cost, shaking the industry., few-shot promptingfew-shot prompting, n.: Showing an AI a handful of examples of a task — inputs with the answers you want — inside the prompt, so it can copy the pattern for a new case., FrontierMathFrontierMath, prop. n.: Several hundred new math problems, kept unpublished so AI systems cannot have trained on them, written by professional mathematicians to check whether machines can do advanced, even research-level, mathematics., GPQAGPQA, prop. n.: A set of very hard multiple-choice science questions written by PhD-level experts, designed so that looking things up online does not help much: capable non-specialists with web access still mostly fail., gpt-ossgpt-oss, prop. n.: Two free-to-download reasoning models released by OpenAI in August 2025 — its first downloadable language models since 2019 — that anyone may run, modify and use commercially on their own hardware., International Mathematical OlympiadInternational Mathematical Olympiad, prop. n.: The world's top math competition for high-school students. Because its problems demand creative written proofs, not just answers, earning a gold medal there became a famous test of AI reasoning, first passed in 2025.