Show HN: We've built a standalone Apple Watch app running LLMs offline, locally

Category: library

Tags: on-device-ai, llm-inference, edge-computing, apple-watch, open-source

Score: 7.5/10 (Innovation: 7, Technical: 8, Documentation: 8, Utility: 7)

NobodyWho is an on-device AI inference engine that enables running large language models locally and offline on various platforms, including pioneering standalone Apple Watch and Vision Pro apps. It stands out for bringing LLM inference to edge devices with GPU acceleration and multi-platform support, combining existing technologies (llama.cpp, GGUF) in a novel cross-platform package.

Target audience: mobile developers, swift developers, kotlin developers, ai engineers, edge device developers

Repository: https://nobodywho.ooo/posts/apple-watch-vision-pro-apps/ · Rust · EUPL-1.2 · 949 stars

View on Hacker News