01Proprietary reasoning models
OpenAI reasoning models
OpenAI's guide to its reasoning models. reasoning.effort accepts none, minimal, low, medium, high, xhigh and max depending on the model, and the GPT-5.6 and GPT-6 families add reasoning.mode (standard or pro). Raw reasoning tokens are not exposed; summaries are opt-in.
Best for: Hard coding, maths and planning tasks via API.
Visit OpenAI reasoning models →
02Proprietary reasoning models
Claude Fable 5.1
Anthropic's model for demanding reasoning and long-horizon agentic work. Adaptive thinking is always on, the default effort is high, with a 1M-token context window and 128K maximum output.
Best for: The hardest reasoning and long-running agent tasks.
Visit Claude Fable 5.1 →
03Proprietary reasoning models
Claude Opus 5.5
Anthropic's model for long-running agentic coding and knowledge work. Adaptive thinking is always on, the default effort is medium, with a 1M-token context window and 128K maximum output.
Best for: Agentic coding and knowledge work.
Visit Claude Opus 5.5 →
04Proprietary reasoning models
Claude Sonnet 5.5
Anthropic's balance of speed and intelligence. It uses adaptive thinking with a default effort of high, a 1M-token context window and 128K maximum output.
Best for: General reasoning with lower latency.
Visit Claude Sonnet 5.5 →
05Proprietary reasoning models
Claude Haiku 5.5
Anthropic's fastest model, for high-volume, latency-sensitive work such as classification, extraction and routing. It uses adaptive thinking with a default effort of medium and a 1M-token context window.
Best for: Fast, high-volume reasoning-light tasks.
Visit Claude Haiku 5.5 →
06Proprietary reasoning models
Gemini 4 Argon
Google's stable flagship reasoning model (gemini-4-argon) for software engineering, maths and science, and multi-step agents. Thinking levels are low, medium and high; the context window is 1,048,576 tokens, with long decode up to 1M output tokens.
Best for: Complex reasoning and agents on Gemini.
Visit Gemini 4 Argon →
07Proprietary reasoning models
Gemini 3.8 Flash
Google's most intelligent Flash model (gemini-3.8-flash), built for long-horizon software engineering and autonomous agents. Thinking levels are low, medium and high; minimal is not supported.
Best for: Faster, lower-cost Gemini reasoning.
Visit Gemini 3.8 Flash →
08Proprietary reasoning models
Gemini 3.8 Live Extended Thinking
Google's audio-to-audio Live API model (gemini-3.8-live-extended-thinking) for voice interactions that need extended background reasoning. Thinking levels are low, medium and high.
Best for: Voice agents that must reason before replying.
Visit Gemini 3.8 Live Extended Thinking →
09Reasoning modes & controls
OpenAI o1
OpenAI's first o-series reasoning model in the API. The deprecations page lists o1 and o1-pro for shutdown on 23 October 2026, with gpt-5.6-sol as the recommended replacement.
Best for: Understanding the o-series lineage and migrating off it.
Visit OpenAI o1 →
10Reasoning modes & controls
OpenAI o3-mini
Smaller o-series reasoning model. The deprecations page lists it for shutdown on 23 October 2026, with gpt-5.6-sol recommended.
Best for: Checking migration paths for low-cost reasoning.
Visit OpenAI o3-mini →
11Reasoning modes & controls
OpenAI o3
Reasoning model for maths, science, coding and visual reasoning, with a 200,000-token context window and 100,000 maximum output tokens. Marked deprecated; the deprecations page lists shutdown on 11 December 2026.
Best for: Reference specifications and migration planning.
Visit OpenAI o3 →
12Reasoning modes & controls
OpenAI o3-pro
Higher-compute variant of o3. Listed for shutdown on 11 December 2026; the deprecations page recommends gpt-5.6-sol with reasoning.mode set to pro.
Best for: Seeing how pro mode replaced a separate model.
Visit OpenAI o3-pro →
13Reasoning modes & controls
OpenAI o4-mini
Compact o-series model that supports detailed reasoning summaries. Listed for shutdown on 23 October 2026, with gpt-5.6-terra recommended.
Best for: Low-cost reasoning migration.
Visit OpenAI o4-mini →
14Reasoning modes & controls
OpenAI deprecations
Shutdown dates and recommended replacements for OpenAI models, including the o-series.
Best for: Checking whether a model is still available.
Visit OpenAI deprecations →
15Reasoning modes & controls
Claude extended thinking (legacy)
Manual extended thinking (thinking.type enabled with budget_tokens, minimum 1,024). Anthropic marks it deprecated on Claude Opus 4.6 and Sonnet 4.6 and not accepted on later models, which use adaptive thinking with an effort setting.
Best for: Migrating older code to adaptive thinking.
Visit Claude extended thinking (legacy) →
16Reasoning modes & controls
Claude thinking overview
How Claude's thinking works: adaptive thinking, the display setting (summarized or omitted) and thinking with tool use. The thinking text returned is a summary, never the raw chain of thought.
Best for: Understanding what Claude's reasoning output contains.
Visit Claude thinking overview →
17Reasoning modes & controls
Claude effort
output_config.effort sets how many tokens Claude spends: low, medium, high, xhigh for long-horizon work and max. Defaults differ by model, for example high on Claude Fable 5.1 and Sonnet 5.5 and medium on Opus 5.5 and Haiku 5.5.
Best for: Trading thoroughness against cost and speed.
Visit Claude effort →
18Reasoning modes & controls
Claude steering thinking and cost
Guidance on letting Claude decide when and how much to think, and on how thinking is priced.
Best for: Tuning thinking depth and cost.
Visit Claude steering thinking and cost →
19Reasoning modes & controls
Gemini thinking (API docs)
thinking_level support varies by model, for example low, medium and high on gemini-4-argon and gemini-3.8-flash. The docs also cover optional thought summaries and encrypted thought signatures that must be returned in stateless mode.
Best for: Setting Gemini thinking levels.
Visit Gemini thinking (API docs) →
20Reasoning modes & controls
Google Cloud: Gemini thinking
Google Cloud documentation on thinking for Gemini models, including thought signatures and a prompting guide.
Best for: Using Gemini thinking on Google Cloud.
Visit Google Cloud: Gemini thinking →
21Reasoning modes & controls
DeepSeek thinking mode (API guide)
Thinking is on by default at high effort and is set with thinking.type or reasoning_effort (low, high, max). The chain of thought comes back in reasoning_content and must be passed back when tools are used.
Best for: Calling DeepSeek's thinking models correctly.
Visit DeepSeek thinking mode (API guide) →
22Reasoning modes & controls
DeepSeek-R1-0528
Updated R1 checkpoint of 685B parameters under the MIT licence. The model card reports AIME 2025 accuracy rising from 70% to 87.5%, which are DeepSeek's own figures.
Best for: Running or studying the updated open R1 weights.
Visit DeepSeek-R1-0528 →
23Reasoning modes & controls
DeepSeek-R1 paper (arXiv)
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.
Best for: How R1 was trained.
Visit DeepSeek-R1 paper (arXiv) →
24Open-weight reasoning models
DeepSeek-R1
DeepSeek's open reasoning model trained with large-scale reinforcement learning, released with distilled smaller versions.
Best for: The reference open reasoning model.
Visit DeepSeek-R1 →
25Open-weight reasoning models
Qwen QwQ-32B
Alibaba's 32-billion-parameter open reasoning model trained with reinforcement learning.
Best for: Capable reasoning at a manageable size.
Visit Qwen QwQ-32B →
26Open-weight reasoning models
Kimi K2
Moonshot AI's open-weight Kimi K2 family, including thinking models aimed at reasoning and agentic tasks.
Best for: Open models for reasoning plus tool use.
Visit Kimi K2 →
27Open-weight reasoning models
Mistral Magistral Small
Mistral AI's Magistral Small reasoning model, with an emphasis on multilingual step-by-step reasoning.
Best for: European and multilingual reasoning.
Visit Mistral Magistral Small →
28Open-weight reasoning models
OpenAI gpt-oss
OpenAI's open-weight reasoning models, released for local and custom deployment.
Best for: Self-hosting a reasoning model with adjustable effort.
Visit OpenAI gpt-oss →
29Open-weight reasoning models
Phi-4-reasoning
Microsoft's small open reasoning model fine-tuned from Phi-4 for maths and science problems. GitHub lists v0.0.9 as the latest release, from January 2026.
Best for: Reasoning on limited hardware.
Visit Phi-4-reasoning →
30Open-weight reasoning models
Llama Nemotron Ultra
NVIDIA's large open reasoning model derived from Llama, with a switchable reasoning mode.
Best for: Enterprise-scale reasoning on NVIDIA GPUs.
Visit Llama Nemotron Ultra →
31Open-weight reasoning models
Olmo 3 Think
Ai2's fully open Olmo 3 Think reasoning model, released with data and training details.
Best for: Studying reasoning with open data and recipes.
Visit Olmo 3 Think →
32Open-weight reasoning models
DeepSeek-V4.1-Flash
DeepSeek's MIT-licensed V4.1-Flash, first published on 10 September 2026. Its model card describes a continuously controllable reasoning effort from 1 to 100.
Best for: Open-weight reasoning with a fine-grained effort control.
Visit DeepSeek-V4.1-Flash →
33Open-weight reasoning models
Kimi-K3
Moonshot AI's Kimi-K3, first published on 13 June 2026, is described on its model card as an open-weight, natively multimodal agentic model of about 2.8 trillion parameters. Its reported results use a reasoning-effort setting.
Best for: Very large open-weight reasoning and agent work.
Visit Kimi-K3 →
34Open-weight reasoning models
GLM-5.3
Z.ai's GLM-5.3, published on 25 August 2026. Its model card says the thinking budget is set with a reasoning_effort parameter with three levels: low, high and max.
Best for: Open-weight models with a three-level thinking budget.
Visit GLM-5.3 →
35Open-weight reasoning models
Qwen3.8-27B
Alibaba's Apache-2.0 Qwen3.8-27B, published in August 2026. Its model card says thinking mode is on by default, can be disabled per request, and reasoning depth can be tuned with reasoning_effort.
Best for: A compact open reasoning model with adjustable depth.
Visit Qwen3.8-27B →
36More reasoning models & provider docs
Learning to reason with LLMs (OpenAI)
OpenAI's announcement of its first reasoning model; the site may block automated checks.
Visit Learning to reason with LLMs (OpenAI) →
37More reasoning models & provider docs
Gemini thinking
Gemini thinking models and budgets.
Visit Gemini thinking →
38More reasoning models & provider docs
Qwen3: Think Deeper, Act Faster
Hybrid thinking and non-thinking models.
Visit Qwen3: Think Deeper, Act Faster →
39More reasoning models & provider docs
Qwen3 Technical Report
Training details for the Qwen3 family.
Visit Qwen3 Technical Report →
40More reasoning models & provider docs
gpt-oss model card
OpenAI open-weight reasoning models.
Visit gpt-oss model card →
41More reasoning models & provider docs
Kimi K2
Open-weight mixture-of-experts model from Moonshot AI.
Visit Kimi K2 →
42More reasoning models & provider docs
Olmo
Fully open models from Ai2 with open data and training code.
Visit Olmo →
43Guides & explainers
Understanding Reasoning LLMs (Sebastian Raschka)
Clear overview of the main ways reasoning models are built.
Visit Understanding Reasoning LLMs (Sebastian Raschka) →
44Guides & explainers
A Visual Guide to Reasoning LLMs (Maarten Grootendorst)
Illustrated walk-through of test-time compute and training.
Visit A Visual Guide to Reasoning LLMs (Maarten Grootendorst) →
45Guides & explainers
Why We Think (Lilian Weng)
Survey of thinking time, test-time compute and chain-of-thought.
Visit Why We Think (Lilian Weng) →
46Guides & explainers
DeepSeek R1's recipe to replicate o1 (Interconnects)
Analysis of the R1 training recipe.
Visit DeepSeek R1's recipe to replicate o1 (Interconnects) →
47Guides & explainers
Chain-of-Thought Prompting (Prompt Engineering Guide)
Practical guide to chain-of-thought prompting.
Visit Chain-of-Thought Prompting (Prompt Engineering Guide) →
48Prompting for reasoning
Show Your Work: Scratchpads for Intermediate Computation
Early work on writing out intermediate steps.
Visit Show Your Work: Scratchpads for Intermediate Computation →
49Prompting for reasoning
Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
The paper that started chain-of-thought prompting.
Visit Chain-of-Thought Prompting Elicits Reasoning in Large Language Models →
50Prompting for reasoning
Large Language Models are Zero-Shot Reasoners
"Let's think step by step" without examples.
Visit Large Language Models are Zero-Shot Reasoners →
51Prompting for reasoning
Self-Consistency Improves Chain of Thought Reasoning
Sample many reasoning paths and take the majority answer.
Visit Self-Consistency Improves Chain of Thought Reasoning →
52Prompting for reasoning
Least-to-Most Prompting
Break problems into easier subproblems.
Visit Least-to-Most Prompting →
53Prompting for reasoning
Tree of Thoughts
Search over branching reasoning paths.
Visit Tree of Thoughts →
54Prompting for reasoning
ReAct: Synergizing Reasoning and Acting
Interleave reasoning with tool use.
Visit ReAct: Synergizing Reasoning and Acting →
55Prompting for reasoning
Towards Reasoning in Large Language Models: A Survey
Early survey of reasoning techniques.
Visit Towards Reasoning in Large Language Models: A Survey →
56Training: verifiers, rewards & RL
STaR: Bootstrapping Reasoning With Reasoning
Learn from self-generated rationales that reach correct answers.
Visit STaR: Bootstrapping Reasoning With Reasoning →
57Training: verifiers, rewards & RL
Training Verifiers to Solve Math Word Problems (GSM8K)
Outcome verifiers and the GSM8K dataset.
Visit Training Verifiers to Solve Math Word Problems (GSM8K) →
58Training: verifiers, rewards & RL
Let's Verify Step by Step
Process supervision beats outcome supervision for maths.
Visit Let's Verify Step by Step →
59Training: verifiers, rewards & RL
Math-Shepherd
Process reward models without human step labels.
Visit Math-Shepherd →
60Training: verifiers, rewards & RL
The Lessons of Developing Process Reward Models in Mathematical Reasoning
Practical lessons and pitfalls with PRMs.
Visit The Lessons of Developing Process Reward Models in Mathematical Reasoning →
61Training: verifiers, rewards & RL
DeepSeekMath
Introduced GRPO, a critic-free RL algorithm widely reused.
Visit DeepSeekMath →
62Training: verifiers, rewards & RL
Kimi k1.5: Scaling Reinforcement Learning with LLMs
RL recipe with long-context scaling.
Visit Kimi k1.5: Scaling Reinforcement Learning with LLMs →
63Training: verifiers, rewards & RL
Tulu 3: Pushing Frontiers in Open Language Model Post-Training
Open post-training recipe including RL with verifiable rewards.
Visit Tulu 3: Pushing Frontiers in Open Language Model Post-Training →
64Training: verifiers, rewards & RL
DAPO: An Open-Source LLM Reinforcement Learning System at Scale
Open RL system with training-stability techniques.
Visit DAPO: An Open-Source LLM Reinforcement Learning System at Scale →
65Training: verifiers, rewards & RL
Demystifying Long Chain-of-Thought Reasoning in LLMs
What makes long reasoning emerge and how to shape it.
Visit Demystifying Long Chain-of-Thought Reasoning in LLMs →
66Training: verifiers, rewards & RL
LLM Post-Training: A Deep Dive into Reasoning Large Language Models
Survey of post-training methods for reasoning.
Visit LLM Post-Training: A Deep Dive into Reasoning Large Language Models →
67Training: verifiers, rewards & RL
Quiet-STaR
Models learn to think before speaking at every token.
Visit Quiet-STaR →
68Training: verifiers, rewards & RL
Training Large Language Models to Reason in a Continuous Latent Space
Reasoning in hidden states instead of words (Coconut).
Visit Training Large Language Models to Reason in a Continuous Latent Space →
69Training: verifiers, rewards & RL
Let's Think Dot by Dot
Hidden computation in filler tokens.
Visit Let's Think Dot by Dot →
70Test-time compute & efficient reasoning
Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
Shows how allocating inference compute can beat bigger models.
Visit Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters →
71Test-time compute & efficient reasoning
s1: Simple test-time scaling
Budget forcing and a small curated reasoning dataset.
Visit s1: Simple test-time scaling →
72Test-time compute & efficient reasoning
From System 1 to System 2: A Survey of Reasoning Large Language Models
Survey of the shift to deliberate reasoning.
Visit From System 1 to System 2: A Survey of Reasoning Large Language Models →
73Test-time compute & efficient reasoning
Towards Reasoning Era: A Survey of Long Chain-of-Thought
Survey of long chain-of-thought reasoning.
Visit Towards Reasoning Era: A Survey of Long Chain-of-Thought →
74Test-time compute & efficient reasoning
Stop Overthinking: A Survey on Efficient Reasoning for LLMs
Methods to shorten reasoning without losing accuracy.
Visit Stop Overthinking: A Survey on Efficient Reasoning for LLMs →
75Test-time compute & efficient reasoning
vLLM reasoning outputs
Serving reasoning models and parsing their thinking content.
Visit vLLM reasoning outputs →
76Test-time compute & efficient reasoning
SGLang documentation
Serving engine with reasoning-model support.
Visit SGLang documentation →
77Open training stacks, libraries & datasets
Open-R1
Hugging Face's open reproduction of the DeepSeek-R1 training pipeline.
Best for: Learning how reasoning models are trained.
Visit Open-R1 →
78Open training stacks, libraries & datasets
s1: Simple test-time scaling
Research code showing that a small reasoning dataset and a budget-forcing technique can improve test-time scaling.
Best for: Understanding test-time compute.
Visit s1: Simple test-time scaling →
79Open training stacks, libraries & datasets
TRL
Hugging Face library for post-training with RL.
Visit TRL →
80Open training stacks, libraries & datasets
verl
Flexible RL training library for LLMs.
Visit verl →
81Open training stacks, libraries & datasets
OpenRLHF
Scalable RLHF and reasoning RL framework.
Visit OpenRLHF →
82Open training stacks, libraries & datasets
open-instruct (Ai2)
Ai2's post-training code, including RL with verifiable rewards.
Visit open-instruct (Ai2) →
83Open training stacks, libraries & datasets
OpenThoughts
Open reasoning datasets and models.
Visit OpenThoughts →
84Open training stacks, libraries & datasets
OpenR1-Math-220k
Reasoning traces for mathematics.
Visit OpenR1-Math-220k →
85Open training stacks, libraries & datasets
NuminaMath 1.5
Competition-level maths problems and solutions.
Visit NuminaMath 1.5 →
86Open training stacks, libraries & datasets
PRM800K
Step-level correctness labels for maths solutions.
Visit PRM800K →
87Benchmarks & evaluation
ARC Prize
Competition and benchmark family built around ARC-AGI puzzles that test skill acquisition on novel tasks.
Best for: Testing generalisation beyond memorised patterns.
Visit ARC Prize →
88Benchmarks & evaluation
Humanity's Last Exam
Multi-subject benchmark of expert-written questions designed to be hard for frontier models.
Best for: Measuring expert-level knowledge and reasoning.
Visit Humanity's Last Exam →
89Benchmarks & evaluation
Epoch AI Benchmarks
Epoch AI's hub that tracks model results across benchmarks over time.
Best for: Comparing models across several benchmarks.
Visit Epoch AI Benchmarks →
90Benchmarks & evaluation
SWE-bench
Benchmark and leaderboard that tests models on resolving real GitHub issues in software repositories.
Best for: Reasoning applied to code.
Visit SWE-bench →
91Benchmarks & evaluation
MATH dataset
Competition mathematics benchmark.
Visit MATH dataset →
92Benchmarks & evaluation
AIME problems (Art of Problem Solving)
Source of the AIME problems used as a benchmark; the site may block automated checks.
Visit AIME problems (Art of Problem Solving) →
93Benchmarks & evaluation
GPQA
Graduate-level science questions.
Visit GPQA →
94Benchmarks & evaluation
MMLU-Pro
Harder multi-task knowledge and reasoning benchmark.
Visit MMLU-Pro →
95Benchmarks & evaluation
LiveCodeBench
Contamination-free code reasoning benchmark.
Visit LiveCodeBench →
96Benchmarks & evaluation
Humanity's Last Exam
Expert-level multi-domain benchmark.
Visit Humanity's Last Exam →
97Benchmarks & evaluation
FrontierMath
Research-level mathematics problems.
Visit FrontierMath →
98Benchmarks & evaluation
simple-evals (OpenAI)
Lightweight evaluation scripts.
Visit simple-evals (OpenAI) →
99Benchmarks & evaluation
lm-evaluation-harness
Standard open evaluation framework.
Visit lm-evaluation-harness →
100Limits, faithfulness & safety
Do NOT Think That Much for 2+3=?
Documents overthinking on simple questions.
Visit Do NOT Think That Much for 2+3=? →
101Limits, faithfulness & safety
The Illusion of Thinking
Studies how reasoning models behave as puzzle complexity rises; its conclusions have been debated.
Visit The Illusion of Thinking →
102Limits, faithfulness & safety
Reasoning Models Don't Always Say What They Think
Chain-of-thought is often not a faithful account of the model's reasoning.
Visit Reasoning Models Don't Always Say What They Think →
103Limits, faithfulness & safety
Chain of Thought Monitorability
Why monitoring reasoning traces is useful for safety and why it may be fragile.
Visit Chain of Thought Monitorability →