mlsanityData to 7 Oct 2026

Introducing LongCat-Flash-Thinking: A Technical Report

arXiv 2509.18883 ↗First posted 23 Sep 2025cs.AIcited by 21

On the map: LLM reasoning → Reinforcement learning for reasoning

also in 2 more
Abstract (arXiv)

We present LongCat-Flash-Thinking, an efficient 560-billion-parameter open-source Mixture-of-Experts (MoE) reasoning model. Its advanced capabilities are cultivated through a meticulously crafted training process, beginning with long Chain-of-Thought (CoT) data cold-start and culminating in large-scale Reinforcement Learning (RL). We first employ a well-designed cold-start training strategy, which significantly enhances the reasoning potential and equips the model with specialized skills in both formal and agentic reasoning. Then, a core innovation is our domain-parallel training scheme, which decouples optimization across distinct domains (e.g., STEM, Code, Agentic) and subsequently fuses the resulting expert models into a single, nearly Pareto-optimal model. This entire process is powered by our Dynamic ORchestration for Asynchronous rollout (DORA) system, a large-scale RL framework that delivers a greater than threefold training speedup over synchronous methods on tens of thousands of accelerators. As a result, LongCat-Flash-Thinking achieves state-of-the-art performance among open-source models on a suite of complex reasoning tasks. The model exhibits exceptional efficiency in agentic reasoning, reducing average token consumption by 64.5% (from 19, 653 to 6, 965) on AIME-25, without degrading task accuracy. We release LongCat-Flash-Thinking to promote further advances in reasoning systems and agentic AI research.

Key facts. Cited by 21 papers in the map (data through 7 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 54 of its 68 references that are in the map. First posted on arXiv 23 Sep 2025.

Builds on

54 of 68 references.

Closest ancestors · 10

Same direction, newest first

  1. 2025Libra: Assessing and Improving Reward Model by Learning to Think29 Jul
  2. 2025AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning30 May
  3. 2025Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?18 Apr
  4. 2025Understanding R1-Zero-Like Training: A Critical Perspective26 Mar
  5. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scale18 Mar
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning22 Jan
  7. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMs22 Jan
  8. 2024Tulu 3: Pushing Frontiers in Open Language Model Post-Training22 Nov
  9. 2023Let's Verify Step by Step31 May
  10. 2017Proximal Policy Optimization Algorithms20 Jul

Other references · 44

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  3. 2025Qwen3 Technical Reportcited by 8,195
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  6. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  7. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  8. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  9. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  10. 2024HybridFlow: A Flexible and Efficient RLHF Frameworkcited by 1,980
  11. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  12. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,880
32 more
  1. 2019Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelismcited by 1,829
  2. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  3. 2023Instruction-Following Evaluation for Large Language Modelscited by 1,143
  4. 2024Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvementcited by 1,057
  5. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 946
  6. 2023TIES-Merging: Resolving Interference When Merging Modelscited by 795
  7. 2023Language Models are Super Mario: Absorbing Abilities from Homologous Models as a Free Lunchcited by 679
  8. 2025τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environmentcited by 558
  9. 2025GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Modelscited by 446
  10. 2021MiniF2F: a cross-system benchmark for formal Olympiad-level mathematicscited by 380
  11. 2025DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decompositioncited by 270
  12. 2024DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Searchcited by 189
  13. 2025ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoningcited by 186
  14. 2024Are We Done with MMLU?cited by 178
  15. 2025Kimina-Prover Preview: Towards Large Formal Reasoning Models with Reinforcement Learningcited by 173
  16. 2025Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learningcited by 159
  17. 2019Mastering Complex Control in MOBA Games with Deep Reinforcement Learningcited by 150
  18. 2023On the Tool Manipulation Capability of Open-source Large Language Modelscited by 122
  19. 2025Seed-Prover: Deep and Broad Reasoning for Automated Theorem Provingcited by 96
  20. 2023Natural Language Reasoning, A Surveycited by 61
  21. 2025StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generationcited by 56
  22. 2024MoE++: Accelerating Mixture-of-Experts Methods with Zero-Computation Expertscited by 55
  23. 2025LongCat-Flash Technical Reportcited by 54
  24. 2024FuseChat: Knowledge Fusion of Chat Modelscited by 53
  25. 2025LlamaRL: A Distributed Asynchronous Reinforcement Learning Framework for Efficient Large-scale LLM Trainingcited by 37
  26. 2024Shortcut-connected Expert Parallelism for Accelerating Mixture-of-Expertscited by 31
  27. 2025OJBench: A Competition Level Code Benchmark For Large Language Modelscited by 22
  28. 2023An Adaptive Placement and Parallelism Framework for Accelerating RLHF Trainingcited by 13
  29. 2025Dynamic Fisher-weighted Model Merging via Bayesian Optimizationcited by 12
  30. 2025SampleMix: A Sample-wise Pre-training Data Mixing Strategey by Coordinating Data Quality and Diversitycited by 10
  31. 2025Do Large Language Models Excel in Complex Logical Reasoning with Formal Language?cited by 4
  32. 2024AntDT: A Self-Adaptive Distributed Training Framework for Leader and Straggler Nodescited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 19.

21 papers.

Sort:

By area:

  1. 2025VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applicationscited by 45
  2. 2026LongCat-Flash-Thinking-2601 Technical Reportcited by 32
  3. 2025Seer: Online Context Learning for Fast Synchronous LLM Reinforcement Learningcited by 21
  4. 2026AgentNoiseBench: Benchmarking Robustness of Tool-Using LLM Agents Under Noisy Conditioncited by 19
  5. 2026DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Trainingcited by 13
  6. 2026XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compressioncited by 6
  7. 2026LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learningcited by 6
  8. 2026LOGIGEN: Logic-Driven Generation of Verifiable Agentic Taskscited by 5
  9. 2026OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyondcited by 5
  10. 2026Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environmentscited by 4
  11. 2026SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipeliningcited by 3
  12. 2026BEAM: Binary Expert Activation Masking for Dynamic Routing in MoEcited by 1
9 more
  1. 2026VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Modelscited by 1
  2. 2026Dynamic Rollout Editing for Reducing Overthinking in RL-Trained Reasoning Modelscited by 1
  3. 2026Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateauscited by 1
  4. 2026Exploring Information Seeking Agent Consolidationcited by 0
  5. 2026XStreamVGGT: Extremely Memory-Efficient Streaming Vision Geometry Grounded Transformer with KV Cache Compressioncited by 0
  6. 2026When to Stop Reusing: Dynamic Gradient Gating for Sample-Efficient RLVRcited by 0
  7. 2026Building Multi-Task Agentic LLMs via Two-Phase Distillationcited by 0
  8. 2026MidTool: Mid-training Data Synthesis for Agentic Tool Usecited by 0
  9. 2026No Pain, More Gain: Iterative Merging for Effective Multi-Teacher On-Policy Distillationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Towards Flash Thinking via Decoupled Advantage Policy Optimization
  2. 2026Mach-Mind-4-Flash Technical Report
  3. 2026Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters
  4. 2026Yuan3.0 Flash: An Open Multimodal Large Language Model for Enterprise Applications
  5. 2026Towards a Mechanistic Understanding of Large Reasoning Models: A Survey of Training, Inference, and Failures
  6. 2025Demystifying Long Chain-of-Thought Reasoning in LLMs
  7. 2024Marco-o1: Towards Open Reasoning Models for Open-Ended Solutions
  8. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Models