使用 Vibe 编码 SwiftUI 应用程序非常有趣
📄 中文摘要
作者最近购买了一台128GB M5 MacBook Pro,发现其在运行本地大型语言模型(LLM)方面表现出色。为了替代 Activity Monitor,作者尝试使用 Vibe 编码开发一些性能监控工具,并对此结果感到满意。这是作者第二次尝试使用 Vibe 编码 macOS 应用程序,之前开发过一个演示应用。通过 Claude Opus 4.6 和 GPT-5.4,作者能够轻松创建 SwiftUI 应用程序,且完整的 SwiftUI 应用可以在一个文本文件中实现,无需打开 Xcode。迄今为止,作者已构建了两个应用程序:Bandwidther 用于监控网络带宽使用情况,Gpuer 用于显示其他相关信息。
📄 English Summary
Vibe coding SwiftUI apps is a lot of fun
The author has recently acquired a 128GB M5 MacBook Pro, which has proven to be very capable of running local large language models (LLMs). Frustrated with Activity Monitor, the author decided to vibe code some alternative performance monitoring tools and is pleased with the results. This marks the author's second experiment with vibe coding macOS apps, following a presentation app developed earlier. Using Claude Opus 4.6 and GPT-5.4, the author can easily create SwiftUI applications, with a complete app fitting in a single text file, eliminating the need to open Xcode. So far, two apps have been built: Bandwidther, which shows network bandwidth usage by applications, and Gpuer, which displays additional relevant information.
Powered by Cloudflare Workers + Payload CMS + Claude 3.5
数据源: OpenAI, Google AI, DeepMind, AWS ML Blog, HuggingFace 等