Chapper: Private AI, supercharged. Run local on-device models, connect LM Studio and more, use MCP tools, Vision, C.A.S.H. shell, JSON Lab, file uploads and more.
Chapper is a powerful, privacy-first AI chat client for iPhone and iPad that runs on-device models and connects directly to local LLMs, LM Studio, self-hosted AI servers, and OpenAI-compatible endpoints.
No accounts. No forced cloud. No subscription.
Whether you run LM Studio, Ollama, llama.cpp, or another OpenAI-compatible endpoint, Chapper gives you a polished, fully native iOS experience for chatting with your models, managing conversations, tuning inference settings, using tools, and working with local AI on your own terms.
Built in SwiftUI from the ground up, Chapper is designed to feel like a real iPhone and iPad app, not a web wrapper.
FEATURES
• On-Device Local Models
Run supported AI models directly on iPhone and iPad for private local inference, including MLX models and Apple Foundation Models where available.
• Personas
Create and switch between custom AI personas, each with their own system prompt, tone, and language. Built-in personas include Code Expert, Translator, Teacher, Analyst, Brainstormer, and more.
• Thought Process
See reasoning output from supported models in a collapsible panel above responses.
• Rich Markdown Rendering
Full support for code blocks, tables, lists, emphasis, and structured responses.
• Model Management
Browse, inspect, select, load, and unload models directly from the app where supported. View quantization, context length, tool support, vision support, and active status.
• MCP Tools
Built-in tool workflows for web search, current time, server status, model listing, chat search, file and directory access, URL fetching, export actions, and more.
• C.A.S.H. Terminal
A built-in sandboxed terminal environment for practical local workflows, file operations, shell scripting, and AI-assisted tool execution.
• Files App Integration
Access Chapper workspace files directly in the iOS Files app, including exports and tool-related content where supported.
• Memory
Save useful long-term information and inject it into conversations automatically for more consistent assistance across chats.
• Prompt Templates
Save and reuse common prompts instantly.
• Full Parameter Control
Adjust temperature, token limits, repeat penalty, Top-K, Top-P, Min-P, TFS-Z, Flash Attention, KV-cache offload, context length, keep-alive, structured output, and more.
• Search
Full-text search across chats and message history.
• Export
Export conversations in multiple formats including TXT, PDF, HTML, Markdown, JSON, CSV, and XML.
• Keychain Support
Store API keys securely using iOS Keychain.
• iCloud Sync
Optionally sync chats across devices with iCloud.
• Analytics Dashboard
View on-device usage stats and activity insights.
• Text to Speech
Listen to responses with native, local, or server-based TTS playback.
• Chat Info
Inspect timestamps, counts, tags, folder assignment, and other conversation metadata.
• Server Tools
Configure, save, and reuse server connections for a smoother local AI workflow.
• JSON Lab
Work with custom JSON requests and experimental API workflows directly inside Chapper.
• Chapper Pro
One-time purchase. Unlock unlimited chat history, all export formats beyond plain text, advanced sampling and structured output controls, MCP tools in normal chat, C.A.S.H. terminal access with shell scripting and AI shell tooling, premium personas and prompt templates, memory and analytics extras, iCloud sync, live vision chat, premium appearance options including additional app icons, and future Pro features. No subscription.
COMPATIBILITY
Chapper works with on-device models, LM Studio, Ollama, llama.cpp, and OpenAI-compatible servers. Some features depend on your model or server.
PRIVACY
Chapper is built with privacy in mind. Your chats stay on your device unless you explicitly enable iCloud sync or connect to your own server. No accounts required. No third-party tracking.
Chapper is built for people who care about local AI, privacy, model control, and a true native iOS experience.
Ratings & Reviews
3.0
out of 5
1 Ratings
Chapper 1.3.0 is a major update focused on local AI, better tool use, and a smoother everyday chat experience.
New in this release:
• On-device local models for private AI directly on iPhone and iPad
• Live Activities for ongoing chats and local generation
• Support for Apple Foundation Models where available
• MLX local model support with improved loading, generation, and model management
• Better local vision model handling
• Improved LM Studio and MCP tool workflows
• New Brave Search option for web search
• More reliable MCP tool execution for files, web search, URLs, time, memory, shell actions, and server tools
• Improved C.A.S.H. filesystem behavior and clearer tool activity reporting
• Updated Pro and local model purchase flows
• UI refinements across Settings, model management, chat, and tool activity views
• More robust iCloud sync behavior
• Improved chat composer behavior and cursor stability
• Fixes for user-reported MCP, local model, StoreKit, and UI issues
• General reliability, compatibility, and polish improvements across Chapper
Chapper 1.3.0 makes local AI, MCP tools, and everyday chat workflows more powerful, private, and reliable.
Version 1.3.0
The developer, Prevolut Ltd, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy .
Data Not Collected
The developer does not collect any data from this app.
Privacy practices may vary, for example, based on the features you use or your age. Learn More
The developer indicated that this app supports the following accessibility features. Learn More
Supported Features
VoiceOver
Voice Control
Larger Text
Dark Interface
Differentiate Without Color Alone
Sufficient Contrast
Information
Seller
Prevolut Ltd
Size
148.5 MB
Category
Productivity
Compatibility
Requires iOS 17.0 or later.
iPhone Requires iOS 17.0 or later.
iPad Requires iPadOS 17.0 or later.
Mac Requires macOS 14.0 or later and a Mac with Apple M1 chip or later.