mlsanityData to 14 Oct 2026

Kimi-Dev: Agentless Training as Skill Prior for SWE-Agents

arXiv 2509.23045 ↗First posted 27 Sep 2025cs.AI cs.CL cs.SEcited by 41

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

Large Language Models (LLMs) are increasingly applied to software engineering (SWE), with SWE-bench as a key benchmark. Solutions are split into SWE-Agent frameworks with multi-turn interactions and workflow-based Agentless methods with single-turn verifiable steps. We argue these paradigms are not mutually exclusive: reasoning-intensive Agentless training induces skill priors, including localization, code edit, and self-reflection that enable efficient and effective SWE-Agent adaptation. In this work, we first curate the Agentless training recipe and present Kimi-Dev, an open-source SWE LLM achieving 60.4% on SWE-bench Verified, the best among workflow approaches. With additional SFT adaptation on 5k publicly-available trajectories, Kimi-Dev powers SWE-Agents to 48.6% pass@1, on par with that of Claude 3.5 Sonnet (241022 version). These results show that structured skill priors from Agentless training can bridge workflow and agentic frameworks for transferable coding agents.

Key facts. Cited by 41 papers in the map (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 24 of its 51 references that are in the map. First posted on arXiv 27 Sep 2025.

Builds on

24 of 51 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Skywork-SWE: Unveiling Data Scaling Laws for Software Engineering in LLMs24 Jun
  2. 2025SWE-rebench: An Automated Pipeline for Task Collection and Decontaminated Evaluation of Software Engineering Agents26 May
  3. 2025SWE-smith: Scaling Data for Software Engineering Agents30 Apr
  4. 2025R2E-Gym: Procedural Environments and Hybrid Verifiers for Scaling Open-Weights SWE Agents9 Apr
  5. 2024Training Software Engineering Agents and Verifiers with SWE-Gym30 Dec
  6. 2024Agentless: Demystifying LLM-based Software Engineering Agents1 Jul

Other references · 18

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2025Qwen3 Technical Reportcited by 8,820
  5. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  6. 2024Qwen2.5 Technical Reportcited by 4,942
  7. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  8. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,373
  9. 2025gpt-oss-120b & gpt-oss-20b Model Cardcited by 1,386
  10. 2024OpenAI o1 System Cardcited by 1,244
  11. 2025Kimi k1.5: Scaling Reinforcement Learning with LLMscited by 1,210
  12. 2025Kimi K2: Open Agentic Intelligencecited by 567
6 more
  1. 2025SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolutioncited by 263
  2. 2025MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attentioncited by 263
  3. 2025Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learningcited by 161
  4. 2025SWE-bench Goes Live!cited by 80
  5. 2025Parallel Scaling Law for Language Modelscited by 30
  6. 2025Data Mixing Can Induce Phase Transitions in Knowledge Acquisitioncited by 16

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 37.

41 papers.

Sort:

By area:

  1. 2025SkyRL-Agent: Efficient RL Training for Multi-turn LLM Agentcited by 48
  2. 2025Toward Training Superintelligent Software Agents through Self-Play SWE-RLcited by 40
  3. 2026SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolvingcited by 31
  4. 2026SWE-World: Building Software Engineering Agents in Docker-Free Environmentscited by 25
  5. 2026SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Trainingcited by 22
  6. 2026Immersion in the GitHub Universe: Scaling Coding Agents to Masterycited by 22
  7. 2026BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?cited by 22
  8. 2026daVinci-Dev: Agent-native Mid-training for Software Engineeringcited by 18
  9. 2025From Code Foundation Models to Agents and Applications: A Comprehensive Survey and Practical Guide to Code Intelligencecited by 15
  10. 2026daVinci-Env: Open SWE Environment Synthesis at Scalecited by 14
  11. 2026From Question Answering to Task Completion: A Survey on Agent System and Harness Designcited by 12
  12. 2026Quantifying and Mitigating Self-Preference Bias of LLM Judgescited by 10
29 more
  1. 2026Dockerless: Environment-Free Program Verifier for Coding Agentscited by 9
  2. 2026Scaling Coding Agents via Atomic Skillscited by 8
  3. 2026Orchard: An Open-Source Agentic Modeling Frameworkcited by 8
  4. 2025Agentic Software Issue Resolution with Large Language Models: A Surveycited by 6
  5. 2026From SWE-ZERO to SWE-HERO: Execution-free to Execution-based Fine-tuning for Software Engineering Agentscited by 6
  6. 2026SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Contextcited by 6
  7. 2026From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AIcited by 6
  8. 2026Hindsight Hint Distillation: Scaffolded Reasoning for SWE Agents from CoT-free Answerscited by 5
  9. 2026Hera: Learning Long-Horizon Coordination for Device-Cloud Collaborative LLM Agentscited by 5
  10. 2026FastContext: Training Efficient Repository Explorer for Coding Agentscited by 5
  11. 2026InCoder-32B-Thinking: Industrial Code World Model for Thinkingcited by 4
  12. 2026Building to the Test: Coding Agents Deliver What You Check, Not What You Requestedcited by 4
  13. 2026SWE-Edit: Rethinking Code Editing for Efficient SWE-Agentcited by 3
  14. 2026PhoenixRepair: Rethinking Repair Strategy Exploration in Software Agentscited by 2
  15. 2026HE-SNR: Uncovering Latent Logic via Entropy for Guiding Mid-Training on SWE-benchcited by 1
  16. 2026M2A: Synergizing Mathematical and Agentic Reasoning in Large Language Modelscited by 1
  17. 2026Change2Task: From Repository Changes to Executable Coding Agent Tasks and Environmentscited by 1
  18. 2026MidTool: Mid-training Data Synthesis for Agentic Tool Usecited by 1
  19. 2026Terminal Agents: A Survey of AI Agents in Command-Line Environmentscited by 1
  20. 2026Signal Reshaping for GRPO in Weak-Feedback Agentic Code Repaircited by 0
  21. 2026DPIAgent: Divide, Protocol, Isolate for Agentic Reproduction Test Generationcited by 0
  22. 2026Towards a Systems Foundation for Agentic Skills: Architecture, Lifecycle, and Securitycited by 0
  23. 2026CodeSkill: Latent Skill Abstraction for Long-Horizon Code Agentscited by 0
  24. 2026Improving Large Language Models for Code through Runtime Program-State Reasoningcited by 0
  25. 2026TrajLong: Co-Designing Agentic and Long-Context Supervision for Mid-Trainingcited by 0
  26. 2026ANTE: Anticipate, Then Execute -- Learning Internal World Models for Coding Agents from Verified Beliefscited by 0
  27. 2026PreciseCoder: Faithful and Generalizable Debugging Beyond Unit-Test Rewardscited by 0
  28. 2026Tev: Scaling Long-Horizon Agentic Environments with Dense Progress Rewardscited by 0
  29. 2026SWE-Stream: Scaling Executable Training Data for SWE Agents via Progressive Construction and Elastic Streamingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
  2. 2026Socratic-SWE: Self-Evolving Coding Agents via Trace-Derived Agent Skills
  3. 2026Yet Even Less Is Even Better For Agentic, Reasoning, and Coding LLMs
  4. 2025Training Versatile Coding Agents in Synthetic Environments
  5. 2025SWE-Dev: Building Software Engineering Agents with Training and Inference Scaling
  6. 2026SWE-Router: Routing in Multi-turn Agentic Software Engineering Tasks
  7. 2026Open-SWE-Traces: Advancing Dual-Mode Multilingual Distillation for Software Engineering Agents
  8. 2026SWE-rebench V2: Language-Agnostic SWE Task Collection at Scale