KeepAI

On-device AI for Android

Your AI.
Your device.
Nothing else.

KeepAI thinks entirely on your phone. Answers, document understanding, image recognition and dictation all happen locally — no cloud, no account, no subscription.

  • 100 % offline
  • Buy once
  • No trackers
  • Two model sizes
€9 once — what you get How it works

Play Store release in preparation. The internal test track is already running.

Why KeepAI

Privacy here is not
a setting.

It is the architecture. There is no server anything could be sent to — the app ships neither analytics nor advertising SDKs.

On-device inference

Every answer is computed locally. After setup the app works in airplane mode.

Two model sizes

Google Gemma 4 as E2B or E4B — pick the one your phone can carry, switchable mid-chat.

Understands documents

Attach PDFs and text, then ask. The content stays in the app's private storage.

Personal Knowledge

Notes, memory entries and chat exports become semantically searchable — fully local.

News

A dated offline snapshot that arrives with every app update — never fetched over the network.

Paper and ink

One light theme, two dark ones and selectable accent colours. Quiet enough for long reads.

A look inside

This is what it looks like.

Chat with an attached PDF and a local answer Choosing between the two local models with storage ledger Switching models directly inside the chat The three modes: Personal, News and Thinking Themes and accent colours Chat in the monochrome dark theme

Engines

Two sizes, one engine.

Both run natively on the device through LiteRT-LM. Download the size your phone can carry and switch whenever you like — even mid-conversation.

Gemma 4 · E2BGemma 4 · E4B
StrengthThe everyday defaultMore headroom on capable phones
RuntimeLiteRT-LMLiteRT-LM
Sizeapprox. 2.4 GBapprox. 3.4 GB
Context32K32K
Memoryfrom 6 GB RAMfrom 8 GB RAM
DeliveryOne-time download during setup — or fully offline via file import.

Your knowledge

What you entrust to KeepAI never leaves your device — not even when it remembers it.

Memory

“Remember that …” — and KeepAI has it ready in every new chat.

Semantic search

Imported PDFs and notes are indexed locally and drawn on while answering.

Delete means delete

Uninstalling removes everything. No copy exists off-device, system backup is disabled.

Price

Buy once.
That's it.

€9One-time purchase · no subscription · no in-app purchases
  • Both engines, all themes, every feature
  • News updates with each app version
  • No ads, no analytics, no accounts
  • Models straight from Hugging Face or via offline import

Price includes VAT, billed through Google Play. Availability and final price may differ by country.

Requirements

What your phone needs.

Recommended

A recent Android device with at least 6 GB RAM — Pixel 8 or newer, Galaxy S23 or newer.

Minimum

Gemma 4 E2B needs 6 GB RAM; below that the app will tell you before downloading.

Storage

Between 2.4 and 3.4 GB free, depending on the size. The app checks before every download.

FAQ

Answered briefly.

Does KeepAI really need no internet?

Only once: to download the AI model during setup. After that the app works fully offline. If you prefer, import the model as a file and never go online at all.

Where are my chats stored?

Exclusively in your device's private app storage. There is no server, no sync and no system backup (allowBackup=false).

Is there a subscription?

No. A one-time purchase of €9, with no further costs and no in-app purchases.

What happens to my voice during dictation?

The recording is transcribed on-device, never stored and never transmitted. The app asks explicitly before first use.

Which models does it run?

Google Gemma 4, as E2B or E4B, via LiteRT-LM. Both are verified by SHA-256 on download.