mlsanityData to 14 Oct 2026

Qwen3 Technical Report

arXiv 2505.09388 ↗First posted 14 May 2025cs.CLcited by 8,820

On the map: LLM reasoning → Reasoning benchmarks and evaluation

also in 1 more
Abstract (arXiv)

In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency, and multilingual capabilities. The Qwen3 series includes models of both dense and Mixture-of-Expert (MoE) architectures, with parameter scales ranging from 0.6 to 235 billion. A key innovation in Qwen3 is the integration of thinking mode (for complex, multi-step reasoning) and non-thinking mode (for rapid, context-driven responses) into a unified framework. This eliminates the need to switch between different models--such as chat-optimized models (e.g., GPT-4o) and dedicated reasoning models (e.g., QwQ-32B)--and enables dynamic mode switching based on user queries or chat templates. Meanwhile, Qwen3 introduces a thinking budget mechanism, allowing users to allocate computational resources adaptively during inference, thereby balancing latency and performance based on task complexity. Moreover, by leveraging the knowledge from the flagship models, we significantly reduce the computational resources required to build smaller-scale models, while ensuring their highly competitive performance. Empirical evaluations demonstrate that Qwen3 achieves state-of-the-art results across diverse benchmarks, including tasks in code generation, mathematical reasoning, agent tasks, etc., competitive against larger MoE models and proprietary models. Compared to its predecessor Qwen2.5, Qwen3 expands multilingual support from 29 to 119 languages and dialects, enhancing global accessibility through improved cross-lingual understanding and generation capabilities. To facilitate reproducibility and community-driven research and development, all Qwen3 models are publicly accessible under Apache 2.0.

Key facts. Cited by 8,820 papers in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 55 of its 72 references that are in the map. First posted on arXiv 14 May 2025.

Builds on

55 of 72 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025AutoLogi: Automated Generation of Logic Puzzles for Evaluating Reasoning Abilities of Large Language Models24 Feb
  2. 2025ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning3 Feb

Other references · 53

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  4. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  7. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  8. 2025Qwen2.5-VL Technical Reportcited by 5,587
  9. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,371
  10. 2024Qwen2.5 Technical Reportcited by 4,942
  11. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  12. 2023Let's Verify Step by Stepcited by 3,789
38 more
  1. 2021Program Synthesis with Large Language Modelscited by 3,637
  2. 2023Qwen Technical Reportcited by 3,344
  3. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
  4. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  5. 2024Qwen2 Technical Reportcited by 2,120
  6. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
  7. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,848
  8. 2025Gemma 3 Technical Reportcited by 1,824
  9. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,807
  10. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,391
  11. 2023GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpointscited by 1,346
  12. 2024RULER: What's the Real Context Size of Your Long-Context Language Models?cited by 1,346
  13. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,228
  14. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  15. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,090
  16. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Trainingcited by 1,053
  17. 2024DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Modelscited by 867
  18. 2024Phi-4 Technical Reportcited by 711
  19. 2023C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Modelscited by 708
  20. 2016Language Modeling with Gated Convolutional Networkscited by 702
  21. 2023Scaling Vision Transformers to 22 Billion Parameterscited by 694
  22. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 601
  23. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  24. 2024From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipelinecited by 500
  25. 2023DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretrainingcited by 399
  26. 2023The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variantscited by 311
  27. 2023Effective Long-Context Scaling of Foundation Modelscited by 303
  28. 2024CRUXEval: A Benchmark for Code Reasoning, Understanding and Executioncited by 300
  29. 2024RegMix: Data Mixture as Regression for Language Model Pre-trainingcited by 177
  30. 2019Neural Machine Translation with Byte-Level Subwordscited by 115
  31. 2025CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratingscited by 115
  32. 2025WritingBench: A Comprehensive Benchmark for Generative Writingcited by 106
  33. 2024Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Followingcited by 102
  34. 2023DoGE: Domain Reweighting with Generalization Estimationcited by 90
  35. 2024INCLUDE: Evaluating Multilingual Language Understanding with Regional Knowledgecited by 86
  36. 2023AlignBench: Benchmarking Chinese Alignment of Large Language Modelscited by 68
  37. 2024Training-Free Long-Context Scaling of Large Language Modelscited by 62
  38. 2025Demons in the Detail: On Implementing Load Balancing Loss for Training Specialized Mixture-of-Expert Modelscited by 54

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2327; 2026: 6493.

8,820 papers.

Sort:

By area:

  1. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,510
  2. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,261
  3. 2025Qwen-Image Technical Reportcited by 1,044
  4. 2026DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligencecited by 1,040
  5. 2026GLM-5: from Vibe Coding to Agentic Engineeringcited by 587
  6. 2025Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoningcited by 559
  7. 2025Group-in-Group Policy Optimization for LLM Agent Trainingcited by 497
  8. 2025Qwen3-Omni Technical Reportcited by 497
  9. 2025GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Modelscited by 467
  10. 2026World Action Models are Zero-shot Policiescited by 465
  11. 2026Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Modelscited by 455
  12. 2026Reinforcement Learning via Self-Distillationcited by 332
38 more
  1. 2026Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipecited by 299
  2. 2025Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformercited by 277
  3. 2025R-Zero: Self-Evolving Reasoning LLM from Zero Datacited by 228
  4. 2025Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergycited by 214
  5. 2025Deep Think with Confidencecited by 205
  6. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  7. 2025BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agentcited by 200
  8. 2025Generalizing Verifiable Instruction Followingcited by 194
  9. 2026GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimizationcited by 193
  10. 2025WebDancer: Towards Autonomous Information Seeking Agencycited by 188
  11. 2026On-Policy Context Distillation for Language Modelscited by 186
  12. 2026Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolationcited by 184
  13. 2025A Survey of Reinforcement Learning for Large Reasoning Modelscited by 180
  14. 2026Entropy-Aware On-Policy Distillation of Language Modelscited by 175
  15. 2026Qwen3.5-Omni Technical Reportcited by 174
  16. 2025Agentic Reinforced Policy Optimizationcited by 172
  17. 2025Skywork Open Reasoner 1 Technical Reportcited by 165
  18. 2025Scaling Latent Reasoning via Looped Language Modelscited by 165
  19. 2026A Survey of On-Policy Distillation for Large Language Modelscited by 163
  20. 2025The Surprising Effectiveness of Negative Reinforcement in LLM Reasoningcited by 162
  21. 2025Qwen3Guard Technical Reportcited by 155
  22. 2025SimpleVLA-RL: Scaling VLA Training via Reinforcement Learningcited by 151
  23. 2026Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixescited by 145
  24. 2025PaddleOCR 3.0 Technical Reportcited by 143
  25. 2026MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agentscited by 142
  26. 2026Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnessescited by 142
  27. 2026Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Groundingcited by 139
  28. 2025ACON: Optimizing Context Compression for Long-horizon LLM Agentscited by 136
  29. 2025OpenRubrics: Towards Scalable Synthetic Rubric Generation for Reward Modeling and LLM Alignmentcited by 132
  30. 2025Emu3.5: Native Multimodal Models are World Learnerscited by 132
  31. 2025SDAR: A Synergistic Diffusion-AutoRegression Paradigm for Scalable Sequence Generationcited by 129
  32. 2026CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verificationcited by 128
  33. 2026VLA-JEPA: Enhancing Vision-Language-Action Model with Latent World Modelcited by 118
  34. 2025Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcingcited by 116
  35. 2025OctoThinker: Mid-training Incentivizes Reinforcement Learning Scalingcited by 113
  36. 2025Reinforcement Learning with Rubric Anchorscited by 113
  37. 2025Geometric-Mean Policy Optimizationcited by 111
  38. 2025MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Serverscited by 109

50 most cited of 8,820 All 8,820 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Qwen3-VL Technical Report
  2. 2025An Empirical Study of Qwen3 Quantization
  3. 2023ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models
  4. 2025Scaling Fine-Grained MoE Beyond 50B Parameters: Empirical Evaluation and Practical Insights
  5. 2025Experts are all you need: A Composable Framework for Large Language Model Inference
  6. 2025dots.llm1 Technical Report
  7. 2024Revolutionizing Database Q&A with Large Language Models: Comprehensive Benchmark and Evaluation
  8. 2024Qwen2-Audio Technical Report