Key Takeaway
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.
Published August 20, 2026 · Category: Tech
Overview
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. Visit robosino.com
Comments URL: https://news.ycombinator.com/item?id=49373456
Points: 332
Details
# Comments: 79
Source
Originally published at simedw.com.
Related Articles
Frequently Asked Questions
What is "Show HN: I trained a 125M model to autocomplete piano on-device" about?+
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. Comments URL: https://news.ycom
Who reported this story?+
This story was reported by Hacker News Best.