# OnDevice LLM - Offline AI Chat — Pay once. Truly private AI.

> Offline AI chat and model execution on iPhone, prioritizing user privacy. (iPhone/iPad app by Mustafa Ergisi.)

- Source: https://appshunter.io/ios/app/ondevice-llm-offline-ai-chat/id6761696880 (this page in markdown: same URL + `.md`)
- Developer: [Mustafa Ergisi](https://appshunter.io/developer/1829311519)
- Category: Utilities, Productivity
- Price: Free with in-app purchases
- Rating: 5.00/5 from 1 App Store ratings
- Age rating: 4+
- Requires: iOS 17.0 · 50 MB
- Languages: American English, Turkish
- Released: 2026-04-16
- Data updated: 2026-08-26
- Monetization: freemium
- User reviews in markdown: https://appshunter.io/ios/app/ondevice-llm-offline-ai-chat/id6761696880/reviews.md

## What is OnDevice LLM?

OnDevice LLM runs powerful open-source AI models 100% on your iPhone — no internet, no cloud, no account.

Your conversations never leave the device. Every word you type, every reply you get, stays on your phone. There is no server to log it, no company to mine it, no account that can be banned.

NEW IN 1.3 — TERNARY BONSAI, HOME SCREEN, DAILY BRIEF
- Ternary Bonsai arrives: 1.7B / 4B / 8B (MLX) — 8B-class quality in just 1.75 GB, fully offline.
- Brand-new Home screen with Quick Actions: ask a document, ask a photo, brief my day, or start a chat.
- Brief my day: combine your Calendar + Reminders with the app's private memory into a morning briefing — fully on-device.
- Better DeepSeek R1 output (fixed chat template).
- Model info cards on every model — check size, fit, and format before loading.

WHAT'S INSIDE
- A curated catalog of top open models:
  - llama.cpp / GGUF: Gemma 4 / Gemma 2, Llama 3.2, Qwen 2.5 / Qwen 3, Phi-4 Mini, SmolLM2, DeepSeek R1.
  - MLX (Apple Silicon native): Ternary Bonsai 1.7B / 4B / 8B.
- Quick Start chips: tap a persona (Translator, Code Helper, Brainstorm, Email Writer, ELI5...) or a demo (Rock Paper Scissors, Click Counter, Snake, Calculator, Tic-Tac-Toe, Drawing Pad, Bouncing Ball, SVG illustration) to send the prompt with one tap.
- Run mini-apps in chat: ask for a game, calculator, drawing pad, or SVG and tap "Run" to use the result right inside the chat — fully offline.
- Markdown chat with code blocks, syntax highlighting, copy + share.
- Attach PDFs, images, or text documents — the AI reads them locally.
- Voice input via on-device speech recognition.
- Persistent chat history with search.
- Personal memory that learns useful facts about you (and stays on your phone).
- Share Extension: send text from any app straight to your AI.

PRIVACY BY DESIGN
- 100% offline inference. Airplane mode works. Cellular off works.
- No account. No email. No tracking SDKs. No analytics that report your prompts.
- Your chats, memories, and downloaded models live in your iPhone's app sandbox only.

WHO IT'S FOR
- Privacy-conscious professionals (legal, medical, journalism, therapy notes).
- Frequent travelers, pilots, mariners — anyone working without reliable internet.
- Users in countries where cloud AI services are blocked.
- Tinkerers and developers who want to load and compare open-source models on a phone.

PRO UPGRADE
- Free plan: 1 model loaded, 3 saved chats.
- Pro: unlimited models, unlimited chats, custom system prompts, custom HuggingFace model URLs.
- One-time Lifetime purchase — buy once, own forever.

Subscriptions auto-renew unless canceled at least 24 hours before the period ends. Manage or cancel any time in Settings - Apple ID - Subscriptions.

Terms of Use (EULA): https://www.apple.com/legal/internet-services/itunes/dev/stdeula/

## Key features

- 100% offline AI inference
- No internet or account needed
- Catalog of open-source AI models
- Analyze documents and images
- Daily briefing with calendar integration
- Run mini-apps within chat
- Persistent chat history and search

## Pricing and in-app purchases

Base price: Free.

**Subscriptions**

| Subscription | Price | Period |
| --- | --- | --- |
| Pro Monthly | $6.99 | per month |
| Pro Yearly | $49.99 | per year |

**One-time purchases**

| Purchase | Price | Type |
| --- | --- | --- |
| Lifetime Pro | $19.99 | one-time |



## Frequently asked questions about OnDevice LLM

### What is OnDevice LLM?

OnDevice LLM is a mobile application that allows you to run powerful AI models entirely on your iPhone or iPad. It operates 100% offline, meaning no internet connection or account is required for its core functionality.

### Does OnDevice LLM have ads?

No, OnDevice LLM does not contain advertisements. The app is designed for a private and uninterrupted user experience, with no tracking SDKs or analytics reporting your prompts.

### What AI models can I use with OnDevice LLM?

The app features a curated catalog of top open-source models, including Gemma, Llama 3.2, Qwen 2.5, Phi-4 Mini, SmolLM2, DeepSeek R1, and Ternary Bonsai models for Apple Silicon.

### How does OnDevice LLM handle privacy?

Privacy is central to OnDevice LLM. All conversations, memories, and downloaded models are stored exclusively within your device's app sandbox. Your data never leaves your phone, ensuring no logging or mining by any company.

### What are the Pro features for OnDevice LLM?

The Pro upgrade offers unlimited model loading and saved chats, along with custom system prompts and the ability to add custom HuggingFace model URLs. This is available as a one-time lifetime purchase.

### Can I use OnDevice LLM with documents or images?

Yes, you can attach PDFs, images, or text documents directly to your chats. The AI will read and process these local files to provide relevant responses.

### What is the 'Brief my day' feature?

The 'Brief my day' feature combines your device's Calendar and Reminders with the app's private memory to generate a personalized morning briefing, all processed securely on your device.

### What devices does OnDevice LLM support?

OnDevice LLM is available for both iPhone and iPad devices, running on Apple Silicon for optimized performance.

## Version history (last 5 releases)

### 1.42 — 2026-08-18

New model: Qwen 3.5 2B.

A generation newer than Qwen 3 at 1.28 GB, so it fits most iPhones. Apache 2.0 licensed and, like everything here, it runs entirely on your device.

Also: models that reason before answering now show a quiet "thinking" indicator while they work, instead of their raw notes.

### 1.41 — 2026-08-12

• Fixed: the app could briefly show the trial-ended screen at launch to customers who had already purchased, before their purchase was confirmed.
• The chat screen now says a model is loading instead of showing an empty state while it opens.

### 1.40 — 2026-08-11

• After your first answer, the chat now offers a next step so you can start a fresh task without losing the current one.

### 1.39 — 2026-08-11

• The chat screen now shows real download progress instead of an empty state, so you can see how long the first model takes.
• You can type your first message while the model is still downloading. It sends by itself the moment the model is ready.

### 1.38 — 2026-08-08

• Fixed: Workspace could stay locked even after subscribing. If you own Lifetime and subscribed to Workspace, it now unlocks correctly.
• Stability improvements.

## Apps similar to OnDevice LLM

| App | Rating | Price | Category |
| --- | --- | --- | --- |
| [Offline AI - Private Chat](https://appshunter.io/ios/app/offline-ai-private-chat/id6760442643) | 4.3 (11) | Free | Utilities |
| [Endemic: Local On-Device AI](https://appshunter.io/ios/app/endemic-local-on-device-ai/id6760927382) | 5.0 (4) | Free | Utilities |
| [Silo - Private AI Assistant](https://appshunter.io/ios/app/silo-private-ai-assistant/id6741248135) | — | Free | Utilities |
| [OfflineGPT - Offline & Private](https://appshunter.io/ios/app/offlinegpt-offline-and-private/id6755918504) | 4.6 (7) | Free | Utilities |
| [MyDeviceAI - Local AI & Search](https://appshunter.io/ios/app/mydeviceai-local-ai-and-search/id6736578281) | 4.9 (14) | Free | Utilities |
| [LM Station-Run AI Offline](https://appshunter.io/ios/app/lm-station-run-ai-offline/id6753220281) | 4.3 (27) | Free | Utilities |

## More apps by Mustafa Ergisi

- [Coffee Budget Tracker](https://appshunter.io/ios/app/coffee-budget-tracker/id6749211840)
- [Prompt Pilot](https://appshunter.io/ios/app/prompt-pilot/id6749590169)
- [AI Word or Not](https://appshunter.io/ios/app/ai-word-or-not/id6749833661)
- [WhenCast: Activity Forecast](https://appshunter.io/ios/app/whencast-activity-forecast/id6751142766)
- [SQL Prep - Learn & Practice](https://appshunter.io/ios/app/sql-prep-learn-and-practice/id6752492649)
- [Real Estate Flyer Maker · CRE](https://appshunter.io/ios/app/real-estate-flyer-maker-cre/id6752876654)

All apps by Mustafa Ergisi: https://appshunter.io/developer/1829311519

## Related topics

[on-device](https://appshunter.io/ios/topics/on-device)

---

*Data collected daily from the US App Store and indexed by [AppsHunter](https://appshunter.io/). User reviews are verbatim App Store reviews. Ratings, prices and chart positions refresh continuously; this snapshot is from 2026-08-26.*
