Packages for keyword “on-device-ml”
These packages are available as a package collection, usable in Xcode or SwiftPM.
Adapt
On-device LoRA personalization for iOS and macOS. Trains, evaluates and rolls back adapters locally — no server, no Python.
ANEEmbed
ANE-optimized CoreML embeddings, converted and quantized from Hugging Face, maintained as a Swift package
SwiftGrad
A tiny autograd engine and neural network library in Swift. Inspired by Karpathy's micrograd.
SwiftRL
On-device reinforcement learning for Swift. Built on SwiftGrad.
AparteSecrets
Finds passwords, API keys, tokens and private keys in text on iOS and macOS: a 196 KB model, offline, no dependency. Swift runtime of aparte-secrets.
AparteTitler
On-device conversation titles for iOS and macOS: a 133 KB model, 17 languages, no network. Swift runtime of aparte-titler.
6 packages.