arXiv:2507.12482cs.SEcs.AI2025-07被引 3

专为调试设计的代码大模型,能高效处理千万行级项目中的复杂错误。

Kodezi Chronos: A Debugging-First Language Model for Repository-Scale Code Understanding

  • 通过图引导检索与持久化调试记忆,实现大规模代码库精准定位
  • 真实场景下修复准确率达67.3%,远超主流模型的14%以下
  • 适合需要深度调试和跨文件分析的研发团队使用

大型语言模型在代码生成和自动化方面取得进展,但在推理时上下文限制和缺乏结构化代码推理能力,导致调试问题仍无法解决。尽管Claude 4.5 Opus在SWE-bench Verified上达到74.40%,Gemini 3 Pro达76.2%,两者在真实多文件调试任务中仍低于20%。我们提出Kodezi Chronos-1,专为调试设计的语言模型,融合自适应图引导检索(支持高达1000万行代码库,92%精度,85%召回率)、基于超过1500万次会话训练的持久化调试记忆,以及七层修复-测试-优化架构。在5000个真实场景中,Chronos-1修复准确率为67.3% ± 2.1%,显著优于Claude 4.1 Opus的14.2% ± 1.3%和GPT-4.1的13.8% ± 1.2%(Cohen's d = 3.87)。在SWE-bench Lite上达到80.33%(241/300)的最高解决率,领先第二名20个百分点,其中在Sympy和Django分别达到96.1%和90.4%的特定项目高分。该模型使调试时间减少40%,迭代次数降低65%,可解决需时序分析的复杂跨文件及跨仓库错误。局限在于硬件相关与动态语言错误。Chronos-1将于2025年Q4在Kodezi OS上线,2026年Q1开放API。

原文摘要 · Abstract (English)

Large Language Models (LLMs) have advanced code generation and software automation but remain constrained by inference-time context and lack structured reasoning over code, leaving debugging largely unsolved. While Claude 4.5 Opus achieves 74.40% on SWE-bench Verified and Gemini 3 Pro reaches 76.2%, both models remain below 20% on real multi-file debugging tasks. We introduce Kodezi Chronos-1, a language model purpose-built for debugging that integrates Adaptive Graph-Guided Retrieval to navigate codebases up to 10 million lines (92% precision, 85% recall), Persistent Debug Memory trained on over 15 million sessions, and a seven-layer fix-test-refine architecture. On 5,000 real-world scenarios, Chronos-1 achieves 67.3% +/- 2.1% fix accuracy compared to 14.2% +/- 1.3% for Claude 4.1 Opus and 13.8% +/- 1.2% for GPT-4.1 (Cohen's d = 3.87). On SWE-bench Lite, Chronos-1 reaches a state-of-the-art 80.33% resolution rate (241 of 300), outperforming the next best system by 20 points and achieving repository-specific highs of 96.1% on Sympy and 90.4% on Django. Chronos-1 reduces debugging time by 40% and iterations by 65%, resolving complex multi-file and cross-repository bugs that require temporal analysis. Limitations remain for hardware-dependent and dynamic language errors, and Chronos-1 will be available in Kodezi OS in Q4 2025 and via API in Q1 2026.

代码调试大模型软件工程智能开发

Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。