leonickson1/Swiftlet

Swiftlet is a Swift and Metal runtime that runs large Qwen Mixture-of-Experts models locally on Apple devices by streaming expert weights from storage, enabling 35B and 80B models to run with low RAM, including on iPhone.

312 best points · 141 comments · 1 Hacker News post · 667 GitHub stars · View on GitHub

Hacker News discussions

• Show HN: Run an 80B Qwen in 4.3 GB of RAM on a Mac, and a 35B on an iPhone (312 points, 141 comments, by leonickson, 2026-08-03)