Kodezi Chronos：面向仓库级代码理解的调试优先语言模型
Kodezi Chronos: A Debugging-First Language Model for Repository-Scale Code Understanding
- Kodezi Inc.（Kodezi公司）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
Kodezi Chronos-1是一种专为调试设计的语言模型，通过自适应图引导检索、持久调试内存和七层架构，在多文件调试任务中达到67.3%的修复准确率，超越现有模型并实现仓库特定的高解决率。
AI中文摘要：
大型语言模型（LLMs）在代码生成和软件自动化方面取得了进展，但受限于推理时间的上下文和缺乏对代码的结构化推理，调试问题仍未得到解决。虽然Claude 4.5 Opus在SWE-bench Verified上达到74.40%，Gemini 3 Pro达到76.2%，但两者在真实多文件调试任务上仍低于20%。我们引入了Kodezi Chronos-1，这是一种专门为调试设计的语言模型，集成了自适应图引导检索以导航多达1000万行的代码库（精度92%，召回率85%），持久调试内存通过超过1500万次会话进行训练，以及七层的修复-测试-细化架构。在5000个真实场景中，Chronos-1在修复准确性上达到67.3%±2.1%，远超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，Chronos-1达到最先进的80.33%的解决率（241/300），超越次优系统20个百分点，并在Sympy和Django上分别达到96.1%和90.4%的仓库特定高分。Chronos-1将调试时间减少了40%，迭代次数减少了65%，解决了需要时间分析的复杂多文件和跨仓库bug。对于硬件依赖性和动态语言错误仍存在限制，Chronos-1将在2025年第四季度的Kodezi OS中发布，并在2026年第一季度通过API发布。
英文摘要：
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.
