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.

语言：Swift · GitHub 星标：665 · Fork：40

话题：local-llm · mixture-of-experts · on-device-ai · qwen · swift

在 GitHub 上查看原仓库