Magenta Real-Time Music Generation Locally on iPhone, Without the GPU
Last Thursday, Deepmind released Magenta Realtime 2 , an open source music generation model. They said it could run on Mac, but not iPhone. As a v̵i̵b̵e̵ ̵c̵o̵d̵i̵n̵g̵ ̵a̵d̵d̵i̵c̵t̵ agentic AI maxxi and person who has melted iPhones before (link at bottom), I took that as a personal challenge and made it my weekend project. On Saturday, I got it to run for 10min straight on an iPhone 12 Pro from 2020 without melting the phone or - shockingly - touching the GPU. How? I chopped the model up into 5 pieces and set them each to run on different parts of Apple's system on a chip (SoC). My past…
Magenta Real-Time Music Generation Locally on iPhone, Without the GPU launched on June 10, 2026 with 9 votes, #1,453 of 22,647 launches that month and more than 75% of that year's launches.
What it does
Google's MRT2 → Core ML, full Apple Neural Engine execution, zero GPU. Runs locally on iPhone without burning your hand. - GitHub - mattmireles/magenta-realtime-2-iphone: Google's MRT2 → Core ML, full Apple Neural Engine execution, zero GPU. Runs locally on iPhone without burning your hand.
A 230M-parameter live-music model sustains its compute and audio-delivery clocks on A17 Pro for ten hot foreground minutes without application GPU execution. A three-seed crossover then localizes an apparent long-horizon quality failure to missing causal decoder history and verifies the repair. A separate no-reset experiment fails its frozen 600-second liveness gate in all three seeds, so the deployed claim remains explicitly bounded. This repository is the public artifact for Google DeepMind's Magenta RealTime 2 ( mrt2_small ) on iPhone: Core ML exporters, frozen fixtures, gate verifiers, machine-readable device receipts, and the paper source. Pre-converted models:…from github.com
In the maker’s words, at launch
Last Thursday, Deepmind released Magenta Realtime 2 , an open source music generation model. They said it could run on Mac, but not iPhone. As a v̵i̵b̵e̵ ̵c̵o̵d̵i̵n̵g̵ ̵a̵d̵d̵i̵c̵t̵ agentic AI maxxi and person who has melted iPhones before (link at bottom), I took that as a personal challenge and made it my weekend…
Does a similar job
all alternatives →
I trained a 125M model to autocomplete piano on-deviceAug 2026 · simedw.com · ▲598I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
Daft Music2025 · daftmusic.app · ▲276Some time ago I moved from Spotify to Apple Music for a few reasons. As someone who builds iOS and macOS apps, I quickly noticed how rough the Apple Music experience felt on Mac. It just wasn’t fun to use. So I decided to challenge myself and see if it was possible to build a full, complex, and performant Apple Music client using SwiftUI. At first, it went surprisingly well. SwiftUI made it easy to build out the main structure and experiment quickly. The UI came together nicely, and it was exciting to see how modern and flexible the framework could be. But the deeper I went, the more cracks…
Run TRELLIS.2 Image-to-3D generation natively on Apple SiliconApr 2026 · github.com · ▲202I ported Microsoft's TRELLIS.2 (4B parameter image-to-3D model) to run on Apple Silicon via PyTorch MPS. The original requires CUDA with flash_attn, nvdiffrast, and custom sparse convolution kernels: none of which work on Mac. I replaced the CUDA-specific ops with pure-PyTorch alternatives: a gather-scatter sparse 3D convolution, SDPA attention for sparse transformers, and a Python-based mesh extraction replacing CUDA hashmap operations. Total changes are a few hundred lines across 9 files. Generates ~400K vertex meshes from single photos in about 3.5 minutes on M4 Pro (24GB). Not as fast as…
Run Stable Diffusion Directly on iPhone2024 · apps.apple.com · ▲162Download Amazing AI by Sindre Sorhus on the App Store. See screenshots, ratings and reviews, user tips, and more apps like Amazing AI.
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 22d ago · stillwet.art
FastRouter.ai▲364Route requests to the right LLM for cost, latency & quality
AI · 23d ago · fastrouter.ai
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, June 2026
the whole month →
Fundraisly▲1,537AI fundraising agent that finds investors and books meetings
AI · Jun 2026 · fundraisly.com
Homebrew 6.0.0▲1,481Today, I’m proud to announce Homebrew 6.0.0. The most significant changes since 5.1.0 are a new tap trust security mechanism, the new faster, smaller, default internal Homebrew JSON API, sandboxing on Linux, better defaults informed by our user survey, many brew bundle improvements, improved performance and initial support for macOS 27 (Golden Gate). Happy to discuss any questions here!
Dev tools · Jun 2026 · brew.sh
Hi HN! I was super frustrated at apps like tiktok and wechat preventing me from saving or downloading the animated gifs, so I made a little tool to turn a screen recording into a slack-sized emoji. Hope you get some value from this!
Life & fun · Jun 2026 · vorpus.github.io
Search 18 years of Hacker News and chart how any topic, tool, or person trended. Overlay terms to compare their rise and fall across 45M posts and comments.
Life & fun · Jun 2026 · hackernewstrends.com