📄 中文摘要
Sweep是一个开源的大型语言模型，拥有15亿参数，专门用于代码编辑的自动补全功能。该模型采用开放权重设计，允许开发者自由访问和修改模型参数。它的主要功能是预测和建议代码的下一步编辑操作，通过分析当前代码上下文来提供智能的编辑建议。这种自动补全功能可以显著提高开发效率，减少重复性编码工作。作为一个开源项目，Sweep不仅为开发者提供了实用的编码辅助工具，还为AI辅助编程领域提供了可供研究和改进的基础模型。
Sweep: 用于下一步编辑自动补全的15亿参数开源模型
出处: Show HN: Sweep, Open-weights 1.5B model for next-edit autocomplete
发布: 2026年1月22日
📄 English Summary
Show HN: Sweep, Open-weights 1.5B model for next-edit autocomplete
Sweep is an open-source large language model with 1.5 billion parameters, specifically designed for code editing autocomplete functionality. The model features open weights architecture, allowing developers to freely access and modify model parameters. Its primary function is to predict and suggest next-step editing operations in code by analyzing the current code context to provide intelligent editing suggestions. This autocomplete functionality can significantly improve development efficiency and reduce repetitive coding work. As an open-source project, Sweep not only provides developers with a practical coding assistance tool but also offers a foundation model for research and improvement in the field of AI-assisted programming.