Pocket AI: offline AI assistant running in your phone, adapted for DeepSeek-R1 (5.37GB)

General Introduction

Pocket AI (PocketPal AI Chinese version) is a powerful offline AI assistant designed to allow users to talk to AI anytime, anywhere. The project is based on Small Language Models (SLMs) and runs on cell phones without an internet connection, especially adapted to Chinese user experience. Pocket AI supports a variety of small language models, including DeepSeek, Danube, etc., with intelligent memory management, real-time performance display and other features. The project currently supports Android devices, and the iOS version is under development, developers are welcome to contribute.

Function List

• offline operation: All AI models run locally without having to worry about network issues.
• Chinese Optimization: Deeply adapted to the Chinese user experience, supporting DeepSeek and other excellent Chinese models.
• simple and easy to use: A well-designed interface that makes AI interaction easy and natural.
• performance optimization: Intelligent memory management with automatic load/unload models.
• Multi-model support: Supports a variety of small language models, including DeepSeek, Danube, and others.
• real time performance: Real-time display of inference speed to keep track of AI response status.
• Equipment Support: Currently supports Android devices, iOS version is under development.

Using Help

Installation process

1. Download the installation package: Download the latest version of the installer via the link below:
   • download link
2. Installation of applications: Once the download is complete, open the installer to install it.
3. Download model files: Select the appropriate model file according to the equipment performance and get it via the following Baidu.com link:
   • Baidu.com Link. link (on a website) Extract code: xtbf
4. Loading ModelsOpen the app, select "Load Local Model", and select the downloaded model file to use it.

Procedure for use

1. Select Model: Enter the Model Configuration page, select the appropriate model according to the device configuration, and click to download the required model.
2. Starting a conversationWait for the model to finish downloading, click the "Load" button to start the dialog with the AI assistant.
3. View History: Left slide to view history, right slide to delete single message, support one click to clear all records.
4. data security: All data is stored locally to ensure security.

Development Configuration

1. Environmental requirements::
   • Node.js (18.0 or later)
   • Yarn
   • React Native CLI
   • Android Studio (Android Development)
   • Xcode (iOS development, macOS only)
2. development step::
   • Cloning code:git clone https://github.com/yzfly/pocketpal-ai-zh.git
   • Switch to the Chinese branch:git checkout chinese
   • Install the dependencies:yarn install
   • Android:yarn android Launching the Android app
   • iOS:yarn ios Launch the iOS emulator
   • Start the development server:yarn start Starting Metro Services

Utility Scripts

• Clean up the build file:yarn clean
• Code check:yarn lint code style checking.yarn typecheck Type checking
• Run the test:yarn test