mlsanityData to 7 Oct 2026

Kimi K2.5: Visual Agentic Intelligence

arXiv 2602.02276 ↗First posted 2 Feb 2026cs.CL cs.AI cs.LGcited by 449

On the map: LLM agents and tool use → Multimodal and embodied agents

Abstract (arXiv)

We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This includes a series of techniques such as joint text-vision pre-training, zero-vision SFT, and joint text-vision reinforcement learning. Building on this multimodal foundation, K2.5 introduces Agent Swarm, a self-directed parallel agent orchestration framework that dynamically decomposes complex tasks into heterogeneous sub-problems and executes them concurrently. Extensive evaluations show that Kimi K2.5 achieves state-of-the-art results across various domains including coding, vision, reasoning, and agentic tasks. Agent Swarm also reduces latency by up to 4.5× over single-agent baselines. We release the post-trained Kimi K2.5 model checkpoint to facilitate future research and real-world applications of agentic intelligence.

Key facts. Cited by 449 papers in the map (data through 7 Oct 2026). Placed in Multimodal and embodied agents, one of 16 directions in LLM agents and tool use. Builds on 61 of its 82 references that are in the map. First posted on arXiv 2 Feb 2026.

Builds on

61 of 82 references.

References · 61

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2017Proximal Policy Optimization Algorithmscited by 14,441
  4. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  5. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  6. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,048
  7. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  8. 2025Qwen3-VL Technical Reportcited by 2,534
  9. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,044
  10. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  11. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 1,906
  12. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
38 more
  1. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  2. 2023MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contextscited by 1,582
  3. 2024Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysiscited by 1,375
  4. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,193
  5. 2024OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environmentscited by 1,134
  6. 2025DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelscited by 755
  7. 2024Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMscited by 719
  8. 2024LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understandingcited by 644
  9. 2023DataComp: In search of the next generation of multimodal datasetscited by 635
  10. 2021Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LMcited by 621
  11. 2025BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agentscited by 616
  12. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 584
  13. 2024BLINK: Multimodal Large Language Models Can See but Not Perceivecited by 529
  14. 2025Kimi K2: Open Agentic Intelligencecited by 523
  15. 2024Measuring Multimodal Mathematical Reasoning with MATH-Vision Datasetcited by 516
  16. 2026Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfacescited by 515
  17. 2024MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmarkcited by 494
  18. 2023OCRBench: On the Hidden Mystery of OCR in Large Multimodal Modelscited by 466
  19. 2024LVBench: An Extreme Long Video Understanding Benchmarkcited by 402
  20. 2025Muon is Scalable for LLM Trainingcited by 381
  21. 2025Kimi-VL Technical Reportcited by 381
  22. 2024LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitaskscited by 367
  23. 2023OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documentscited by 316
  24. 2025PaperBench: Evaluating AI's Ability to Replicate AI Researchcited by 286
  25. 2025SWE-Bench Pro: Can AI Agents Solve Long-Horizon Software Engineering Tasks?cited by 254
  26. 2024CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMscited by 229
  27. 2023Patch n' Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolutioncited by 226
  28. 2025Video-MMMU: Evaluating Knowledge Acquisition from Multi-Discipline Professional Videoscited by 225
  29. 2023Multimodal C4: An Open, Billion-scale Corpus of Images Interleaved with Textcited by 208
  30. 2024SciCode: A Research Coding Benchmark Curated by Scientistscited by 172
  31. 2025OpenCUA: Open Foundations for Computer-Use Agentscited by 151
  32. 2024OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotationscited by 144
  33. 2025GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Taskscited by 142
  34. 2025ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarizationcited by 132
  35. 2025MMVU: Measuring Expert-Level Multi-Discipline Video Understandingcited by 131
  36. 2025MotionBench: Benchmarking and Improving Fine-grained Video Motion Understanding for Vision Language Modelscited by 69
  37. 2025Towards Robust Mathematical Reasoningcited by 67
  38. 2025WideSearch: Benchmarking Agentic Broad Info-Seekingcited by 66

50 most cited of 61 All 61 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 449.

449 papers.

Sort:

By area:

  1. 2026GLM-5: from Vibe Coding to Agentic Engineeringcited by 502
  2. 2026Gemma 4 Technical Reportcited by 187
  3. 2026Kimi K3: Open Frontier Intelligencecited by 99
  4. 2026Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillationcited by 71
  5. 2026SkillRouter: Skill Routing for LLM Agents at Scalecited by 57
  6. 2026Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agentscited by 57
  7. 2026Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perceptioncited by 47
  8. 2026Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understandingcited by 46
  9. 2026REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agentscited by 43
  10. 2026Uni-OPD: Unifying On-Policy Distillation with a Dual-Perspective Recipecited by 43
  11. 2026MetaClaw: Just Talk -- An Agent That Meta-Learns and Evolves in the Wildcited by 35
  12. 2026OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Datacited by 32
38 more
  1. 2026LongCat-Next: Lexicalizing Modalities as Discrete Tokenscited by 30
  2. 2026ClawBench: Can AI Agents Complete Everyday Online Tasks?cited by 28
  3. 2026Lightning OPD: Efficient Post-Training for Large Reasoning Models with Offline On-Policy Distillationcited by 28
  4. 2026ClawGuard: A Runtime Security Framework for Tool-Augmented LLM Agents Against Indirect Prompt Injectioncited by 25
  5. 2026Gym-Anything: Turn any Software into an Agent Environmentcited by 24
  6. 2026AgentVista: Evaluating Multimodal Agents in Ultra-Challenging Realistic Visual Scenarioscited by 22
  7. 2026Qwen-AgentWorld: Language World Models for General Agentscited by 22
  8. 2026BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?cited by 20
  9. 2026Composer 2 Technical Reportcited by 20
  10. 2026Agentic Aggregation for Parallel Scaling of Long-Horizon Agentic Taskscited by 18
  11. 2026GEMS: Agent-Native Multimodal Generation with Memory and Skillscited by 16
  12. 2026Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agentcited by 16
  13. 2026Uncertainty Quantification in LLM Agents: Foundations, Emerging Challenges, and Opportunitiescited by 15
  14. 2026HunyuanOCR-1.5: Making Lightweight OCR VLMs Faster and Bettercited by 14
  15. 2026PERMA: Benchmarking Personalized Memory Agents via Event-Driven Preference and Realistic Task Environmentscited by 13
  16. 2026Continuous Latent Diffusion Language Modelcited by 12
  17. 2026Memex(RL): Scaling Long-Horizon LLM Agents via Indexed Experience Memorycited by 11
  18. 2026Synthesizing Multi-Agent Harnesses for Vulnerability Discoverycited by 11
  19. 2026Vero: An Open RL Recipe for General Visual Reasoningcited by 10
  20. 2026ToolCUA: Towards Optimal GUI-Tool Path Orchestration for Computer Use Agentscited by 10
  21. 2026GenClaw: Code-Driven Agentic Image Generationcited by 10
  22. 2026Agent-as-a-Router: Agentic Model Routing for Coding Taskscited by 10
  23. 2026WideSeek-R1: Exploring Width Scaling for Broad Information Seeking via Multi-Agent Reinforcement Learningcited by 9
  24. 2026Trust Your Critic: Robust Reward Modeling and Reinforcement Learning for Faithful Image Editing and Generationcited by 9
  25. 2026AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agentscited by 9
  26. 2026AeSlides: Incentivizing Aesthetic Layout in LLM-Based Slide Generation via Verifiable Rewardscited by 9
  27. 2026Can Coding Agents Reproduce Findings in Computational Materials Science?cited by 9
  28. 2026WebVR: Benchmarking Multimodal LLMs for WebPage Recreation from Videos via Human-Aligned Visual Rubricscited by 8
  29. 2026VisBrowse-Bench: Benchmarking Visual-Native Search for Multimodal Browsing Agentscited by 8
  30. 2026Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decodingcited by 8
  31. 2026HyperEyes: Dual-Grained Efficiency-Aware Reinforcement Learning for Parallel Multimodal Search Agentscited by 8
  32. 2026Auditing Agent Harness Safetycited by 8
  33. 2026QUEST: Training Frontier Deep Research Agents with Fully Synthetic Taskscited by 8
  34. 2026Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retentioncited by 8
  35. 2026Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Applicationcited by 8
  36. 2026Dockerless: Environment-Free Program Verifier for Coding Agentscited by 8
  37. 2026EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfercited by 8
  38. 2026AgentAbstain: Do LLM Agents Know When Not to Act?cited by 8

50 most cited of 449 All 449 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Agent-X: Evaluating Deep Multimodal Reasoning in Vision-Centric Agentic Tasks
  2. 2026GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents
  3. 2025DeepEyesV2: Toward Agentic Multimodal Model
  4. 2025Skywork-R1V4: Toward Agentic Multimodal Intelligence through Interleaved Thinking with Images and DeepResearch
  5. 2025SenseNova-MARS: Empowering Multimodal Agentic Reasoning and Search via Reinforcement Learning
  6. 2026GenAgent: Scaling Text-to-Image Generation via Agentic Multimodal Reasoning
  7. 2024Optimus-1: Hybrid Multimodal Memory Empowered Agents Excel in Long-Horizon Tasks
  8. 2026Gym-V: A Unified Vision Environment System for Agentic Vision Research