Reasoning model
A reasoning model is a machine-learning model designed or trained to perform multistep analysis or inference before producing an answer or result.
Reinforcement Learning from Human Feedback
A reasoning language model is a language model optimized to solve problems through explicit or implicit multi-step reasoning before producing a final answer.
- How it is used
- Uses reasoning models to improve performance on verifiable tasks and exploit additional computation during generation.
- How deep it goes
- Explains the RLVR training loop, common training choices, earlier methods, and evaluation changes.
- Examples
- Uses STaR, VinePPO, Tülu 3, and an MMLU multiple-choice prompt to illustrate training and evaluation.
- What you can do after
- Evaluate and understand reasoning-model training choices and generation-based evaluation formats.
Read how Reinforcement Learning from Human Feedback explains reasoning model →
Build a Reasoning Model (From Scratch)
A language model optimized to deliberate over complex problems, often producing intermediate steps or using tools to improve answer quality.
- How it is used
- Deploys intermediate-token generation for complex mathematics, coding, logic, planning, and tool-selection tasks.
- How deep it goes
- Defines visible and hidden intermediate steps, contrasts pattern matching with explicit logic, and outlines inference and training methods without implementing them.
- Examples
- Uses contradictory bird premises, chain-of-thought diagrams, token-by-token generation, roadmap figures, and math-verifier workflows.
- What you can do after
- Evaluate/understand when intermediate reasoning improves accuracy and how verification can measure or train reasoning behavior.
Read how Build a Reasoning Model (From Scratch) explains reasoning model →
AI Agents in Action, Second Edition
A reasoning model is a language model that explicitly allocates computation to multistep deliberation before producing its final output.
- How it is used
- Supports planning, tool selection, and checking progress in agents carrying out multistep goals.
- How deep it goes
- Defines explicit deliberation, contrasts it with single-pass output, and gives effort-setting guidance without detailing model internals.
- Examples
- Uses a travel-planning goal and a planning agent with a sequential thinking server to illustrate agent workflows.
- What you can do after
- Operate/configure reasoning effort for routine and complex agent steps, and limit agent iterations.
Read how AI Agents in Action, Second Edition explains reasoning model →
Machines that Think
A reasoning model is a language model designed and trained to perform intermediate reasoning computation before producing an answer.
- How it is used
- Uses reasoning models to examine how intermediate computation affects apparent language-model abilities and their evaluation.
- How deep it goes
- Explains training and inference-time behavior, offers possible mechanisms, and examines limitations without settling why intermediate reasoning works.
- Examples
- Uses OpenAI o1, DeepSeek-R1, and controlled puzzles including Tower of Hanoi and River Crossing as examples.
- What you can do after
- Evaluate reasoning-model performance by distinguishing visible explanations from intermediate computation and considering alternative causes of benchmark failure.
LLM Customization and Fine-Tuning
A model optimized to perform multi-step reasoning tasks, especially those requiring structured deliberation.
- How it is used
- Selects models for complex multi-step tasks and as teachers when distilling reasoning behavior.
- How deep it goes
- Defines intermediate reasoning and gives selection and generation constraints, but does not detail model training.
- Examples
- Names OpenAI o-series planners and DeepSeek-R1-style teachers for math, code, and planning.
- What you can do after
- Apply the approach by selecting a reasoning-specialist teacher, budgeting longer outputs, filtering reasoning traces, and evaluating intermediate steps.
Read how LLM Customization and Fine-Tuning explains reasoning model →
Introduction to Generative AI, Second Edition
A reasoning model is a language model that generates intermediate reasoning steps before producing a final answer, especially for difficult problems.
- How it is used
- Uses intermediate analysis and self-critique before answering constrained questions.
- How deep it goes
- Explains the observable reasoning sequence but stops short of disclosing training specifics.
- Examples
- Illustrates rhyme construction through planning, brainstorming, self-correction, and final poem generation.
- What you can do after
- Evaluate how intermediate reasoning differs from direct answer generation.
Read how Introduction to Generative AI, Second Edition explains reasoning model →
Context Engineering
A language model optimized to solve demanding problems by generating intermediate reasoning steps before a final output.
- How it is used
- Deploy reasoning models for demanding, multi-step problem solving within LLM systems.
- How deep it goes
- Defines intermediate reasoning tokens and their relation to next-token prediction, without detailing the underlying mechanism further.
- Examples
- Names mathematics, coding, and planning as tasks suited to reasoning models.
- What you can do after
- Apply reasoning models by specifying the task, grounding data, operational constraints, examples, and tool definitions.
Also covered in
- Build a DeepSeek Model (From Scratch)In early access (MEAP)
- Build an AI Agent (From Scratch)
- AI Agents and Applications
- The Art of AI Product Development
- Building Reliable AI Systems
What to know first
- DeepSeek-R1DeepSeek-R1 is a reasoning-focused large language model developed by DeepSeek.
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- reinforcement learningReinforcement learning is a machine-learning paradigm in which an agent learns behavior or a decision policy through interaction with an environment and evaluative feedback, such as rewards.
- AI agentAn AI agent is a computational system that uses artificial intelligence to interpret information and select or perform actions toward goals while interacting with an environment.
Related concepts
- DeepSeek-R1includes · DeepSeek-R1 is a reasoning-focused large language model developed by DeepSeek.
- large language modelkind of · A large language model is a machine-learning model trained on extensive text data to process and generate natural language.
- language modelkind of · A language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- chain-of-thought promptingcontrasts with · Chain-of-thought prompting is a technique that elicits or supplies intermediate reasoning steps to guide a model toward a final answer.
- chain-of-thoughtuses · A chain-of-thought is an ordered sequence of intermediate reasoning steps that connects information or a problem to a conclusion, whether expressed or generated as text.
Questions about reasoning model
What is reasoning model?
A reasoning model is a machine-learning model designed or trained to perform multistep analysis or inference before producing an answer or result.
Which Manning books explain reasoning model?
Reinforcement Learning from Human Feedback, Build a Reasoning Model (From Scratch), AI Agents in Action, Second Edition, Machines that Think, LLM Customization and Fine-Tuning, Introduction to Generative AI, Second Edition and Context Engineering.
What should I know before learning reasoning model?
Manning books that teach reasoning model usually introduce DeepSeek-R1, language model, reinforcement learning and AI agent first.
What is the difference between reasoning model and chain-of-thought prompting?
A reasoning model is a machine-learning model designed or trained to perform multistep analysis or inference before producing an answer or result. Chain-of-thought prompting is a technique that elicits or supplies intermediate reasoning steps to guide a model toward a final answer.
What kinds of reasoning model are there?
Kinds of reasoning model include DeepSeek-R1.
What is reasoning model a kind of?
Reasoning model is a kind of large language model. A large language model is a machine-learning model trained on extensive text data to process and generate natural language.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.






