I trained a 125M model to autocomplete piano on-device
I 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.
In plain words
This is a free app that uses a 125-million-parameter transformer model to autocomplete piano performances in real time on iPhones. Users play a few notes on a MIDI piano, and the model continues the melody entirely on-device, processing around 108 notes per second on an iPhone 15. It works similarly to code autocomplete tools but for musical performance. The developer trained the model using aggressive data cleaning and direct preference optimization, and made it available for anyone to try.
written from the facts on this page · September 2026
From the sources
Can a small transformer autocomplete MIDI performances in real time?
TL;DR: I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The biggest improvements came from finding the right MIDI representation, cleaning the training data aggressively, and adding DPO post-training. Almost a year ago, I started tinkering with an idea: connect my MIDI piano to my phone, play something, and have AI autocomplete the song for me. Think GitHub Copilot, but for piano. It turned out to be a deeper rabbit hole than I expected. Fourteen experiments later, it is finally at a point where I am happy enough with it to write about. The app, RollTab, is available for free here if you have a MIDI keyboard and an…from simedw.com
More ai this month
the category →
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 17d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 9d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
