ListAIse
Get appDownload
Skip to main content

Use On-Device AI

Install the local models, choose which supported requests stay on your phone, and understand storage, warm-up, offline, and fallback behavior.

Last updated: 2026-09-05

9 min read

Local Processing
For supported requests
Offline-Capable
After model installation
Local Responses
After model warm-up
Local Voice Model
Installed separately

What is On-Device AI?

ListAIse offers two AI modes: Cloud AI (default) and On-Device AI. Cloud AI sends the request to a remote service for processing. On-Device AI uses the installed local model for supported tasks and keeps that request on your phone.

Cloud AI

  • No storage needed
  • Always up-to-date model
  • Requires internet
  • Data processed remotely

On-Device AI

  • Supported requests stay local
  • Supported tasks work offline
  • No network round trip
  • Requires 3 to 4 GB of storage

Assistant and substitutions use cloud AI

The Assistant and ingredient substitutions require a connection even when On-Device is selected. The app asks for cloud consent for the session before sending these requests. Choosing On-Device does not disable normal account or shared-list synchronization.

Which AI am I using right now?

Open any list, add an item by voice, and look at the small label below the item while it processes. It shows either "On-device (Gemma)" or "Cloud" to tell you exactly which AI is active.

How to Enable On-Device AI

Select On-Device AI in Settings, then download the model. The model remains on your device after installation.

1

Open App Settings

Tap the profile icon in the top-right corner of the home screen, then select Settings.
2

Navigate to AI Provider

Scroll to the AI section and tap AI Provider. You'll see the current selection: Cloud (default) or On-Device.
3

Select On-Device AI

Tap On-Device (Gemma). ListAIse will prompt you to download the model if it's not already on your device. Confirm to begin the download, which is approximately 3 to 4 GB.
4

Wait for Download

The model downloads in the background. You can continue using the app normally during this time. A progress indicator shows download status. Pause keeps the partial download so you can resume later; Cancel removes it. If a download fails, use Retry.
5

You're ready

Once installed and selected, the local model processes supported AI tasks on your device. The Assistant and ingredient substitutions continue to require cloud processing and consent.

Wi-Fi recommended

The Gemma model is 3 to 4 GB. Download it over Wi-Fi to avoid using mobile data.

First-Run Warm-Up

The first time you use On-Device AI after launching the app, the model needs to load into memory. This is called the warm-up. It can happen again after the app releases the model, including after more than two minutes in the background.

Allow time for the model to load

Loading and processing time depend on your device and the request. A loaded model avoids another initial load and a network round trip, but responses still take time to generate.

What you'll see during warm-up

1

Loading sheet appears

A modal sheet slides up showing "Preparing on-device AI…" with an indeterminate progress indicator. ListAIse is loading the Gemma model into memory.
2

Model ready

Once loaded, the sheet closes and your request is processed. The model may need to load again after being released from memory.

Proactive warm-up

ListAIse tries to warm up the model in the background when you open a list, before you tap the microphone. This can reduce the wait for the first request.

On-Device Voice Transcription

On-Device AI includes a separate model for voice transcription: Whisper, running locally via a compact on-device implementation. This means your voice never leaves your phone, including for transcription.

Whisper On-Device

The Whisper model handles speech-to-text entirely on your device. Once installed, voice input works without an internet connection, including in stores with poor signal.

Setting up on-device voice

1

Tap the microphone icon

In any list, tap the microphone button to start voice input.
2

Install the voice model

If Whisper isn't installed yet, a sheet appears showing the model size and an Install button. Tap it to download.
3

Download completes

A progress bar shows the download. Once complete, voice input activates immediately. Future sessions use the already-installed model.

Model sizes

Whisper is available in multiple sizes (approximately 600 MB to 1.5 GB). Larger models offer better accuracy across accents and noisy environments. Choose based on your available storage and accuracy needs.

Privacy and offline use

On-Device AI keeps supported requests local and can run without a connection after the models are installed.

Local processing

Supported requests handled by the installed local model are processed on your device. Shared-list synchronization and any cloud processing you choose still use online services.

Fully Offline

Supported local AI tasks work without a connection once the required models are installed. The Assistant, ingredient substitutions, and fetching recipe URLs still need internet access.

No network round trip

After the initial warm-up, requests are processed locally without waiting for a server response.

You're in Control

You decide which AI handles your data. Switch between Cloud and On-Device anytime from Settings. Your preference persists across sessions.

Device Requirements

On-Device AI runs large neural network models, so it requires available storage space and a modern device processor.

ModelPurposeSize
Gemma 4Recipe analysis, AI features3.1 GB to 4.4 GB
WhisperVoice transcriptionAbout 600 MB to 1.5 GB

Storage tip

You don't need to install both models. Download Gemma for AI features or Whisper for voice input. Each model can be enabled or disabled independently.

Fallback Behavior

If On-Device AI encounters an error (model loading failed, not enough memory, etc.), ListAIse never silently falls back to cloud AI. You always stay in control.

When local AI fails, you choose

  • Use cloud this time processes this request through Cloud AI without changing your AI provider setting.
  • Cancel stops the request without sending the data.

Cloud fallback requires confirmation

ListAIse asks before sending a failed local request to Cloud AI. The fallback sheet lets you use the cloud for that request or cancel it.

Best Practices

Choose the mode that fits your connection, storage, and processing preference.

Use On-Device AI when…

  • • You're shopping in areas with poor connectivity
  • • You want supported AI requests to be processed locally
  • • You use voice input frequently and want offline transcription
  • • You want supported requests to avoid network delays

Use Cloud AI when…

  • • Your device has limited storage (under 5 GB free)
  • • You're on a newer device that hasn't warmed up the model yet
  • • You want to conserve device RAM during heavy multitasking

First session tip

Before shopping without a connection, install the required models and try the local task you intend to use. Allow time for the model to load again if the app has been in the background.

Use these guides in ListAIse

Download the app to follow the steps on this page.

App StoreGoogle Play

Was this page helpful?

Your feedback helps us decide what to improve next.