HN Showcase
Back to browse
GitHub Repository

Swift package for running LiteRT-LM models on iOS. Wraps Google's C API in a clean, async/await Swift interface.

43 starsSwift

Running Gemma 4 on an iPhone 13 Pro

by dengjiuhong·Apr 15, 2026·1 point·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidNiche GemShip It

Clean Swift wrapper for Gemma 4 with vision and audio on iPhone.

Strengths
  • •Wraps Google's C API into idiomatic async/await Swift interface.
  • •Supports text, vision, and audio modalities with streaming.
  • •Includes working entitlement setup for increased memory limit.
Weaknesses
  • •Requires iPhone 13 Pro+ with 6GB RAM, limiting device compatibility.
  • •Library wrapper rather than novel architecture or complete product.
Category
AI/ML
Target Audience

iOS developers building apps with local LLMs

Similar To

MLX · llama.cpp · Off Grid

Post Description

I just open-sourced how https://github.com/mylovelycodes/LiteRTLM-Swift

LiteRTLM-Swift lets you run LLMs locally with a clean Swift API. - On-device inference - No cloud required - Built for iOS

Similar Projects

AI/ML●●●Banger

SwiftLM – Qwen Chat on iPhone, 100B+ Moe on M5 Pro 64GB (Native Swift)

Native Swift inference with SSD streaming runs 100B MoE models without kernel panics.

WizardryNiche Gem
aegis_camera
122mo ago
AI/ML●●●Banger

iPhone ANE holds LLM tok/s while MLX and LiteRT thermal-throttle

LiteRT beats MLX on Gemma memory while CoreML sips power on the Neural Engine.

Dark HorseBig BrainSolve My Problem
mlboy
1014d ago