Skip to content

Your reading, your data

KMReader keeps every Komga login, preference, and cache on your device. No account, no analytics, nothing sold.

KMReader keeps every Komga login, preference, and cache on your device. There is no KMReader account, no analytics, and nothing is sold or shared.

What stays on your device

KMReader stores the following locally so you can use several servers, load quickly, and keep reading offline:

  • Komga server addresses, display names, and sign-in credentials — an app-managed API key when the server supports one, otherwise your username and password
  • Reading progress, cached pages and covers, and downloaded books
  • Preferences like themes, layouts, and cache budgets

We cannot view or retrieve any of it. KMReader does not use iCloud sync, analytics, or crash reporting.

How local storage works

Everything lives in KMReader’s app sandbox, which iOS and macOS encrypt at rest as part of standard data protection. Removing a server profile deletes its saved credentials, cached pages, covers, and downloaded books.

You control when to clear caches or sign out. Uninstalling the app removes everything.

Network connections

  • Your Komga servers. KMReader talks directly to the servers you add, using HTTPS when your server supports it. We never proxy, log, or inspect that traffic.
  • Manga translation model packs. When you download an OCR model pack, KMReader fetches the catalog and the pack from models.kmworks.date. These are plain file downloads: no account, library content, or reading data is sent, though the host sees the usual request details such as your IP address.
  • Your translation provider, if you configure one (see below).
  • Apple services you choose to use, such as in-app purchases through the App Store, Handoff, Spotlight, and widgets, are handled by Apple under Apple’s privacy policy.

Manga translation

Text detection and OCR run entirely on your device — page images are never uploaded, to us or to anyone else.

If you use Apple’s built-in translation, the recognized text stays on your device too. If you configure a custom AI translation provider, only the recognized text is sent, over your own API key, to the endpoint you chose. With Include Comic Metadata turned on (it is off by default), the series and chapter titles, authors, and genres are added to that request. That traffic goes straight from your device to your provider; we run no servers in between.

No third-party tracking

The app ships without advertising SDKs, analytics beacons, social trackers, or data brokers. What happens in KMReader stays with you.

Managing or deleting data

Because everything lives on-device, you can:

  • Remove a server profile to delete its credentials, caches, and downloads
  • Use the in-app cache controls to clear pages, books, or covers
  • Delete the app to remove every remaining file and preference

If you reinstall KMReader, you start with a clean slate.

Updates to this policy

We occasionally refine this document to reflect new features or legal requirements. When that happens, we update the date below and publish the latest version on this page.

Questions?

Reach out anytime on GitHub or email us.

Last updated: October 2026

Navigation

Type to search…

↑↓ navigate↵ selectEsc close