# InferrLM — On-device AI

> On-device AI language models for offline, private chat and analysis. (iPhone/iPad app by Subhajit Gorai.)

- Source: https://appshunter.io/ios/app/inferrlm/id6754396856 (this page in markdown: same URL + `.md`)
- Developer: [Subhajit Gorai](https://appshunter.io/developer/1848200873)
- Category: Utilities, Developer Tools
- Price: Free
- Rating: 5.00/5 from 1 App Store ratings
- Age rating: 4+
- Requires: iOS 26.0 · 124 MB
- Languages: English
- Released: 2025-11-19
- Data updated: 2026-07-02
- Monetization: free
- User reviews in markdown: https://appshunter.io/ios/app/inferrlm/id6754396856/reviews.md

## What is InferrLM?

Run AI language models directly on your phone, offline. Chat with AI without internet, keep your conversations private, and access advanced features like document analysis and image recognition.

Who Is This For?

Developers who want to experiment with language models locally. Privacy-conscious users who don't want their data in the cloud. Students who need AI help but don't always have internet. Anyone curious about running real AI on their phone.

Key Features:

- Run AI models locally on your device - no internet needed
- Complete privacy - your chats never leave your phone
- Support for vision models - analyze images and documents
- Built-in OCR to extract text from photos and PDFs
- Optional cloud AI support (ChatGPT, Claude, Gemini, DeepSeek) with your own API keys
- Download optimized models designed for mobile devices
- RAG (Retrieval-Augmented Generation) for smarter document understanding
- Share your AI assistant across devices on your WiFi network

## Key features

- Run AI models locally, no internet required
- Complete conversation privacy
- Analyze images and documents
- Extract text from photos and PDFs
- Optional cloud AI integration
- RAG for enhanced document understanding


## Frequently asked questions about InferrLM

### What AI models can I run with InferrLM?

InferrLM allows you to run AI models locally on your device. It also offers optional cloud AI support for services like ChatGPT, Claude, Gemini, and DeepSeek, provided you have your own API keys.

### Does InferrLM require an internet connection?

No, InferrLM is designed to run AI models directly on your device, meaning you can chat with AI and use its features completely offline without an internet connection.

### How does InferrLM ensure user privacy?

Your conversations and data remain entirely on your phone. InferrLM prioritizes complete privacy, ensuring that your chats never leave your device when using the local AI models.

### Can InferrLM analyze images and documents?

Yes, InferrLM supports vision models for analyzing images and documents. It also includes built-in OCR functionality to extract text from photos and PDFs.

### What devices is InferrLM available on?

InferrLM is available for iPhone and iPad devices, allowing users to experience on-device AI capabilities on their Apple mobile devices.

### How often is InferrLM updated?

The latest version of InferrLM is 0.8.7, and it was last updated on March 30, 2026. This indicates a commitment to ongoing development and improvements.

### What is the age rating for InferrLM?

InferrLM has an age rating of 4+, making it suitable for a wide range of users, including younger individuals interested in AI technology.

### Does InferrLM have ads?

No, InferrLM does not contain advertisements. As a free application, it provides an ad-free experience for all users.

## Version history (last 5 releases)

### 0.8.7 — 2026-03-30

- Fixed critical crashes
- Fixed model export bug
- Improved thinking parser for reasoning models
- Added sliders for model parameters settings
- Removed default system prompt

### 0.8.5 — 2026-03-22

- Added Qwen 3.5 support for MLX
- Improved server logs
- Bug fixes

### 0.8.4 — 2026-03-19

- Added OpenAI-compatible API endpoints (/v1/chat/completions, /v1/models)
- Fixed connection issues with PC clients
- Improved server stability and error handling

### 0.8.1 — 2026-03-10

- Authentication hotfix

### 0.8.0 — 2026-03-09

- InferrLM now supports MLX. Download MLX supported models by going into Models > Download Models > Search on HuggingFace (e.g. "Qwen 3.5 MLX").
- Underlying Llama.cpp engine is now updated, hence support for newer GGUF models.
- Faster chat performace - UI is optimized.
- Improved storage manager - faster model read & write performance.
- Model initialization runtime settings (context window, batch size, GPU offloading etc.) support for GGUF models in model selector.
- Branching support for chat.
- Math & LaTeX rendering in chat.
- API cloning support for remote models (DeepSeek support has now been dropped).
- Fixed random app crashes.
- Many other improvements & bug fixes.

## Apps similar to InferrLM

| App | Rating | Price | Category |
| --- | --- | --- | --- |
| [Kai - Grocery List](https://appshunter.io/ios/app/kai-grocery-list/id6747010480) | — | Free | Utilities |
| [Chinny: Offline Voice Cloner](https://appshunter.io/ios/app/chinny-offline-voice-cloner/id6753816417) | 3.4 (5) | Free | Utilities |
| [VisionLib Companion](https://appshunter.io/ios/app/visionlib-companion/id1464664198) | — | Free | Utilities |
| [Expo AI](https://appshunter.io/ios/app/expo-ai/id6739983829) | 5.0 (1) | Free | Utilities |
| [Bielaruskaja Lacinka](https://appshunter.io/ios/app/bielaruskaja-lacinka/id1635188101) | 5.0 (1) | Free | Utilities |
| [PocketServer: Folder Sharing](https://appshunter.io/ios/app/pocketserver-folder-sharing/id6743850070) | 5.0 (19) | Free | Utilities |

---

*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 2026-07-02.*