Preview
Description
You kick off a training run. An egg appears on your lock screen. While your model learns, the egg wobbles, glows, and cracks — happy when loss is falling, nervous when it spikes, sleepy when things plateau. When the run finishes, it hatches, and you meet what you made: a one-of-a-kind pixel creature whose personality comes from how the training actually went. If the run crashes, well… you'll want to be more careful with the next egg. Lossy is a training monitor with a heartbeat. Under the cute is a real tool: - Loss curve and progress at a glance, on OLED black - Live Activities: hatch progress on your lock screen and Dynamic Island - Widgets: your incubator on the home screen - Real-time push when a run finishes or crashes — no more tab-refreshing - A shelf of every creature you've ever hatched, each one a training run you can look up Setup is two lines: pip install lossy import lossy run = lossy.init(project="my-project") lossy.log({"train/loss": loss}, step=step) Pair once with a token from the app. No account, no API keys, no analytics — and Unpair erases everything server-side. Your metrics belong to you; we just make them adorable.
Technical Analysis
Package Overview Basic metadata
Bundle ID
com.benrio.lossy Current Version 1.5.0
Minimum iOS 17.0
Package Size App Store Size
Package Metadata 0 / 0 version records
Detailed binary, privacy manifest, framework, and Info.plist results have not been indexed for this package yet. The verified App Store and archive metadata above remains available.
Known Version Records
0 version records with stored package metadata
Stored metadata does not guarantee a package is currently available to download.
| Details | Version / IDs | Version date | Action |
|---|