- Reasoning models are trained through reinforcement learning on tasks with verifiable answers like competition mathematics and coding challenges.
- OpenAI o1 achieved 83% on AIME math versus 13% for GPT-4, with o3 reaching 96.7% on AIME 2025.
- Major reasoning models include OpenAI o1/o3, Anthropic Claude 3.7 Sonnet, DeepSeek R1, and Google Gemini 2.5 Pro.
- Use reasoning models for hard mathematical problems, competitive programming, and multi-step logical deduction where thinking more helps.
- Don't use reasoning models for simple tasks like Q&A or creative writing due to added cost and latency without justified benefits.
What Changed with Reasoning Models
Standard LLMs generate text tokenTokenThe basic unit of text that an LLM processes — roughly corresponding to a word or word-piece. Models read input and produce output in tokens, which is also how API usage is measured and billed.Learn more → by token without any explicit planningPlanningAn agent's decomposition of a goal into an ordered set of steps or subgoals before or while acting to achieve complex objectives systematically.Learn more → phase. They commit to their first token and build from there — which works brilliantly for most tasks but fails on problems requiring multi-step deduction, backtracking, or careful error correction. The model has no way to reconsider an early token once it's generated. This is why Chain of ThoughtChain of ThoughtA prompting technique that encourages an LLM to reason step by step before giving a final answer, dramatically improving performance on complex tasks.Learn more → prompting was developed as a workaround — forcing the Foundation ModelFoundation ModelA large AI model trained on broad data at massive scale that can be adapted to a wide variety of downstream tasks, forming the 'foundation' for specialized applications.Learn more → to externalise intermediate steps before committing to an answer.
Reasoning models allocate additional compute at inferenceInferenceThe process of running a trained AI model to generate outputs — what happens when you send a prompt and receive a response.Learn more → time by generating a 'thinking' token sequence before producing a final answer. During this thinking phase, the model explores approaches, catches errors in its own logic, tries alternative paths, and builds up to the answer incrementally. This is qualitatively different from simply generating a longer response. Reasoning models are also the backbone of AgentAgentAn LLM-powered system that can take actions, use tools, and pursue multi-step goals autonomously without human input at each step.Learn more → systems — where an AI must plan, use tools, and verify its own outputs across many steps.
How Reasoning Models Are Trained
Reasoning models are trained primarily through reinforcement learning on tasks with verifiable answers: competition mathematics, formal logic puzzles, algorithmic coding challenges. The model receives a signal only when it reaches the correct final answer — not for the quality of its intermediate reasoning. Over thousands of training steps, it discovers reasoning strategies that reliably produce correct answers.
This outcome-based training is key. Rather than being taught how to reason by human annotators (who might reinforce suboptimal reasoning patterns), the model discovers what works through trial and error at scale. The resulting reasoning chains often look different from human step-by-step working — they reflect what the model found effective, not what a human would explicitly write.
Key Reasoning Models
OpenAI o1 (September 2024) was the breakthrough: 83% on AIMEAIMEAIME (American Invitational Mathematics Examination) problems are challenging high-school competition math questions used to evaluate advanced mathematical reasoning capabilities in AI models.Learn more → math vs 13% for GPTGPTGenerative Pre-trained Transformer — the model architecture and family name behind OpenAI's most famous models, from GPT-2 to GPT-5.Learn more →-4. o3 (January 2025) improved further: 87.7% on GPQAGPQAA graduate-level, 'Google-proof' multiple-choice benchmark of hard science questions designed to resist easy lookup and test deep reasoning.Learn more → Diamond (near human expert level), 96.7% on AIME 2025. o4-mini offers strong reasoning at a lower price point. Anthropic's Claude 3.7 Sonnet with extended thinking is competitive and uniquely shows users the thinking process.
DeepSeek R1 demonstrated that reasoning modelReasoning ModelA class of LLMs trained specifically to 'think' through problems step by step using extended chain-of-thought reasoning before producing a final answer, excelling at math, coding, and complex logic.Learn more → training can be done outside OpenAI, matching o1 as an open-weightOpen-WeightA model whose trained weights are publicly available for download, allowing anyone to run, fine-tune, or build on top of it — distinct from fully open-source (which also includes training code and data).Learn more → model. Google's Gemini 2.5 Pro includes deep thinking capabilities. Mistral's Magistral Medium is a European-developed reasoning model. The technique has spread rapidly across the field.
When to Use Reasoning Models
Use reasoning models for: hard mathematical problems, competitive programming, multi-step logical deduction, scientific analysis with complex inference, legal reasoning, architectural decisions that require exploring many options. The performance gap over standard models is largest for tasks where 'thinking more helps' — problems where the answer isn't immediately obvious.
Don't use reasoning models for: simple Q&A, creative writing, summarization, straightforward coding, conversational applications. The added cost (both money and latencyLatencyThe delay between sending a request to an LLM and receiving the first token of the response, often measured as Time to First Token (TTFT).Learn more →) isn't justified when standard models perform adequately. A practical routing strategy: try the cheapest capable model first, escalate to reasoning if the output fails quality checks.