Skip to main content
PromptQuorumBuilt for humans. Structured for AI.
Home/Power Local LLM/Local AI Chat Pocket Review: Offline LLM and Image Chat for iPhone and iPad
Mobile & Edge LLMs

Local AI Chat Pocket Review: Offline LLM and Image Chat for iPhone and iPad

·8 min read·By Hans Kuepper · Founder of PromptQuorum · Discovery engine for open-weight & open-source AI
Read in:
🇺🇸en🇩🇪de🇫🇷fr🇯🇵ja🇨🇳zh🇪🇸es🇧🇷pt🇸🇦ar🇰🇷ko

Warning: Local AI Chat - pocket has had no update since 27 May 2026 (version 0.5.6), so it may be unmaintained — check that it works on your device before relying on it. Local AI Chat - pocket is a freemium, closed-source iOS and iPadOS app that downloads open models such as Llama, Gemma, Qwen and Mistral to the device, generates images, and chats offline. Per its App Store listing it needs iOS or iPadOS 18.6 or later and is free to download, with weekly, monthly and yearly in-app purchases. This review covers version 0.5.6 and is based on the public listing, product site and privacy policy, not hands-on testing.

Local AI Chat - pocket, which its own site calls Pocket, is an iPhone and iPad app that downloads open language models to the device, generates images, and chats with no connection. This review covers what its App Store listing, product site and privacy policy document, what is unclear, and how it compares with other iOS local-AI apps.

Local AI Chat Pocket Review: Offline LLM and Image Chat for iPhone and iPad

Key Takeaways

  • Version reviewed: 0.5.6, as shown on the App Store listing.
  • Price: free to download; in-app purchases are listed at $6.99 weekly, $14.99 monthly and $99.99 yearly.
  • Platform: iOS and iPadOS 18.6 or later; the listing also shows visionOS 2.6 or later.
  • Standout features: broad model list, image generation, Shortcuts integration, editable system prompt.
  • Openness: closed source; no public repository was found.

📍 In One Sentence

Local AI Chat - pocket, by developer Rica Harz, is a freemium iPhone and iPad app that downloads open language models to the device, generates images, and chats offline.

💬 In Plain Terms

Think of it as a pocket-sized offline chatbot: once a model is on the phone, no connection is needed, and the app adds Shortcuts and image generation on top.

Get Local AI Chat Pocket

Local AI Chat - pocket is distributed through the Apple App Store. The download is 256.4 MB and the app is a free download with in-app purchases.

This review is a companion to PromptQuorum's Local LLM Software Directory, which lists the app alongside other mobile and desktop local AI tools.

Channel
Get It
Apple App StoreLocal AI Chat - pocket on the App Store
Websiteoneai.framer.website
SupportPocket support page
Privacy policyPocket privacy policy

No public source repository was found. There is no Android or Mac build; the listing covers iPhone, iPad and Apple Vision only.

Local AI Chat Pocket at a Glance

Attribute
Local AI Chat Pocket
PlatformiOS, iPadOS (18.6+)
PriceFree, in-app purchases
LicenseClosed source
Runs fully offlineYes, once a model is on the device
In-app model downloadsNot stated in the listing
Image inputNot stated in the listing
Image generationYes, on-device image models
Voice input / outputNot stated in the listing

Attributes follow the mobile-chat comparison used in the Local LLM Software Directory. They come from the App Store listing and product site, which do not cover every feature.

What Local AI Chat Pocket Is

Local AI Chat - pocket is an offline chat app built around switchable on-device models. The listing pitches "the simplest way to use private LLMs" that "works fully offline", and its product site calls the app Pocket, "the local LLM app that keeps your data private".

It is published by Rica Harz, an individual developer, and is categorized as Productivity. Do not confuse it with similarly named apps such as "Private AI Chat: Pocket LLM", which has a different developer.

This review draws on the App Store listing, the product site and the privacy policy only. It does not include hands-on testing, so speed, stability and answer quality are not rated here.

How to Get Started

Setup is an App Store install, then picking a model inside the app. The listing does not document the first-run flow, so the steps below follow what it states.

  1. 1
    Check device and OS
    Why it matters: The listing requires iOS or iPadOS 18.6 or later, or visionOS 2.6 or later.
  2. 2
    Install the app
    Why it matters: Get it from the [App Store](https://apps.apple.com/us/app/local-ai-chat-pocket/id6476791661); the download is 256.4 MB before any model.
  3. 3
    Choose a model
    Why it matters: The listing says you can choose a new model or switch between existing ones in the app; larger models need more of the device's memory.
  4. 4
    Chat offline, then customize
    Why it matters: Once a model is on the device, chat works with no connection. Adjust the theme, fonts and system prompt, or call the model from Shortcuts.

Features

The app's selling point is a wide model list in a simple interface. Everything below comes from the App Store listing.

  • Language models. The listing names Llama 3.2 1B and 3B, Llama 3 8B, Mistral 7B, Phi, Gemma 2 2B and 9B, OpenELM, Qwen 3 and Deepseek, mostly as 4-bit builds.
  • Image generation. Image models are supported alongside text models, with a switch between them; the listing does not name which image models.
  • Shortcuts. You can chat with the local model or send its output to other actions through Apple Shortcuts.
  • Personalization. Theme, fonts and the system prompt can be adjusted in the app, and chats can be pinned.
  • Languages. The interface is available in English plus 22 other languages.

Device Requirements

The app requires iOS or iPadOS 18.6 or later, or visionOS 2.6 or later. The listing gives no RAM figure and no list of supported chips.

Memory is the real limit: a 1B model fits far more devices than a 7B or 9B one. Neither the listing nor the site says which models fit which iPhone, so test a small model first.

Pricing and Privacy

Local AI Chat - pocket is free to download, with in-app purchases listed at $6.99 weekly, $14.99 monthly and $99.99 yearly. Neither the listing nor the site says what the paid tiers unlock, so check the paywall in the app before relying on a feature.

The listing says "zero data collection" and that your data never leaves the device. The privacy policy, effective February 17, 2026, is more detailed: it describes minimal anonymized telemetry, crash and performance data kept for up to 12 months, and says chats and personal content stay on the device by default.

The policy also describes an optional cloud processing mode, off by default, that would send text, images and chats to the third-party provider TogetherAI. The policy does not cover subscription payments. The app is closed source and this review did not inspect network traffic, so these statements are the developer's own.

Trade-Offs: Benefits vs. Limitations

Offline chat

What it means in real use:
Chat with downloaded models with no connection.
Limitation / caveat:
Needs iOS 18.6 and enough memory for the chosen model.

Wide model list

What it means in real use:
Llama, Gemma, Qwen, Mistral, Phi and Deepseek in one app.
Limitation / caveat:
No sizes, fit guidance or runtime are published per device.

Image generation

What it means in real use:
Text and image models in the same app.
Limitation / caveat:
The listing does not name the image models or speeds.

Shortcuts support

What it means in real use:
Call the local model from your own automations.
Limitation / caveat:
Only what the listing states; no action list is published.

Free to start

What it means in real use:
No cost to download and try.
Limitation / caveat:
Paid tiers cost up to $99.99 a year with no stated feature split.

Local AI Chat Pocket vs. Alternatives

Local AI Chat Pocket

Platforms:
iOS, iPadOS
Price and license:
Freemium, closed source
Model source:
In-app model list
Key difference:
Adds image generation and Shortcuts; weekly paid tier

Locally AI

Platforms:
iOS, Mac
Price and license:
Freemium, closed source
Model source:
In-app model catalog
Key difference:
Also on Mac; no image generation is listed
Articles about Locally AI (4)

Also mentioned in:

Private Mind

Platforms:
iOS, Android
Price and license:
Free, MIT
Model source:
Hugging Face downloads
Key difference:
Open source and on Android too; fully free
Articles about Private Mind (1)

Also mentioned in:

Oscilla

Platforms:
iOS
Price and license:
Free, closed source
Model source:
40+ in-app models
Key difference:
Larger fixed catalog; no in-app purchases listed
Articles about Oscilla (1)

PocketPal AI

Platforms:
iOS, Android
Price and license:
Free, MIT
Model source:
GGUF models
Key difference:
Open source, llama.cpp-based, cross-platform
Articles about PocketPal AI (9)

+24 more not shown

Platform, price, and feature details for third-party apps change frequently. Verify current specifics on each app's own listing before deciding.

Compare in directory

Who Should Use Local AI Chat Pocket

  • People who want one offline app for text and image models. Both are listed in the same interface.
  • Shortcuts users. The Shortcuts integration lets other automations call the local model.
  • Anyone who wants a broad model list to try. Llama, Gemma, Qwen, Mistral and more are named.

Who Should Not Use Local AI Chat Pocket

  • Anyone who needs open source. The code is not public; Private Mind and PocketPal AI are open source.
  • Android or Mac users. Only iPhone, iPad and Apple Vision are listed; PocketPal AI and Private Mind both run on Android.
  • Devices below iOS or iPadOS 18.6. The listing sets that floor.
  • Anyone who needs a strict no-telemetry guarantee. The privacy policy describes anonymous telemetry and an optional cloud mode; read it first.
  • Anyone who wants a recently updated app. Version 0.5.6 dates from May 27, 2026, with release notes reading only "Bug fixes and performance improvements."

Frequently Asked Questions

Is Local AI Chat Pocket free?

It is free to download with in-app purchases at $6.99 weekly, $14.99 monthly or $99.99 yearly, as shown on the App Store listing.

Who makes Local AI Chat - pocket?

Rica Harz, an individual developer, per the App Store and the privacy policy. Apps with similar names, such as "Private AI Chat: Pocket LLM", come from different developers.

Can it run without an internet connection?

Yes, once a model is on the device. The optional cloud mode in the privacy policy is off by default.

Does it work on Android or Mac?

No. The listing covers iPhone, iPad and Apple Vision only.

Which models can I run?

The listing names Llama, Gemma, Qwen, Mistral, Phi, OpenELM and Deepseek models, mostly 4-bit builds, plus image models.

Verdict

Local AI Chat - pocket is a good fit on paper for someone who wants a simple offline chat app with a wide model list, image generation and Shortcuts on an iPhone or iPad running iOS 18.6 or later.

The caveats are transparency, pricing and upkeep: it is closed source, the paid tiers are undocumented, the privacy policy is more nuanced than the listing's "zero data collection", and the last update was in May 2026.

If you want open source, start with Private Mind. If you want image generation and Shortcuts in one app, try the free tier before paying.

Sources

← Back to Power Local LLM

MCP