Private, offline AI for Android

An AI that runs on your phone.

Lite Mind is a free Android app that runs a language model on the phone itself. Type a question, or tap the microphone and say it out loud. Either way it works the answer out on the device, offline, and nothing you say leaves it.

Free, no ads, nothing to sign up for. Android 8.0 and up.

Lite Mind's chat screen: the fox mark above the greeting "Late one. What's on your mind?" and an "Ask anything" box.
On device
The model runs on your phone
Offline
After the first download
No account
Nothing to sign up for
Free
No ads, no subscription

Two ways to ask.

Typing or talking. Both run offline, on the phone.

Offline chat, with no connection

Pick a model and download it once. After that the app works in aeroplane mode, on a train, anywhere. Replies arrive as the model writes them and the chat follows along. Your history is kept on the phone.

  • Proper markdown: headings, lists, tables and links
  • Code blocks with a copy button
  • Chats saved on the device, yours to delete
A Lite Mind conversation: the question "whats the difference between RAM and storage" and the model's written answer below it.

Speak instead of typing

Tap the microphone, talk, tap stop. whisper.cpp turns the recording into text on the phone and drops it into the message box, where you can edit it before sending. Your voice never leaves the device either.

  • A 57 MB voice model, fetched the first time you tap the microphone
  • Microphone permission asked at that tap, never at launch
  • Tap anywhere to stop; the text lands in the box, not in the chat
Lite Mind listening: a pulsing orange circle above the word "Listening" and the hint "Say what is on your mind. Tap anywhere when you are done."

Two small language models to choose from.

Both are small enough to sit on a phone. The app fetches them from Hugging Face, and you can switch between them whenever you like.

Lite Mind's model picker: "Pick a model to start", with Fast at 0.6 GB marked in use and Smarter at 0.7 GB offering a download button.

Fast Default

Answers in a second or two and runs comfortably on older phones. Good for everyday questions, short writing and quick explanations.

  • Model Qwen3.5-0.8B
  • Download 0.6 GB
  • Needs 2 GB RAM
  • Licence Apache 2.0

Smarter

Handles harder questions and writes longer, more accurate answers. Built for phones, so it stays quick despite being the larger of the two.

  • Model LFM2.5-1.2B Instruct
  • Download 0.7 GB, plus a 0.2 GB draft
  • Needs 3 GB RAM
  • Licence LFM Open License v1.0

Getting started

  1. Install it

    The app opens straight into the chat screen. There is no account to make and no setup wizard to sit through.

  2. Download a model

    Pick Fast or Smarter from the sheet that sits over the chat. It is a few hundred megabytes, fetched once over your connection.

  3. Turn the connection off

    From then on it works in aeroplane mode, on a train, or anywhere else you happen to be. The model is already on your phone.

There is nowhere for your conversation to go. The app has no server behind it, and no privacy setting you need to hunt for and switch on.

Nothing is collected

No accounts, no analytics, no crash reporting to anyone else. Your chats and settings live on the phone, and uninstalling takes them with it.

What the network is for

Fetching a model from Hugging Face over HTTPS, and asking Google Play whether a newer version exists. Never for answering you.

Dictation too

whisper.cpp transcribes on the device. No recording is uploaded, and the microphone is only asked for when you tap it.

Read the full privacy policy →

Questions about running AI offline

Does it really work with no connection?

Yes, once a model is downloaded. That download is the only part that needs the internet. After it finishes you can switch the connection off and carry on.

Is this an offline version of ChatGPT?

No. ChatGPT runs on OpenAI's servers and needs a connection. Lite Mind is a separate app that runs an open model on your own phone, so it answers offline. The models are far smaller, so the answers are shorter and simpler.

Which phones can run it?

Android 8.0 or newer on a modern ARM phone, which covers nearly everything sold since about 2017. Fast wants 2 GB of RAM and Smarter 3 GB, plus room for the model file itself.

How much storage does it need?

One model, so 0.6 GB for Fast or 0.7 GB for Smarter, and Smarter adds a 0.2 GB draft file. The 57 MB voice model is only fetched if you use dictation. Deleting a model gives the space straight back.

How good are the answers?

These are small models, roughly a thousandth the size of the ones behind the big cloud assistants. They do well on everyday questions, explanations and short writing. They also get things wrong, so check anything that matters.

Does it cost anything?

No. There is no price, no subscription, no advertising and nothing to unlock.

Keep your thoughts on your phone.

Free, no ads, nothing to sign up for. Download it once and it works anywhere.