ShellGPT is a free, multi-platform desktop client for interacting with large language models like ChatGPT, GPT4, NewBing, and Claude, along with AI image generation. It targets users seeking an integrated, no-login, no-key experience for advanced AI chat, translation, and image creation, offering a convenient alternative to web interfaces.

How It Works

The application aggregates access to various AI models, including ChatGPT, GPT4, NewBing, and Claude, providing a unified interface. It features automatic node speed testing and switching for optimal performance, supports long-form text translation without word limits, and integrates AI image generation capabilities. A key architectural advantage is its plugin system, allowing users to extend functionality by adding custom nodes or features.

Quick Start & Requirements

• Install: Download pre-compiled binaries from the releases page.
• Prerequisites: No API keys or logins are required for core functionality. Supports Windows, macOS, Linux, and Android.
• Setup: Minimal setup required; download and run.
• Online Use: Online Demo 1 and Online Demo 2.

Highlighted Details

• Supports GPT3.5, GPT4, NewBing (with internet search), and Claude.
• Features a plugin system for extensibility.
• Offers multiple modes: Dialogue, Translation, Long Text Translation, Text Polishing.
• Includes chat history saving and management.

Maintenance & Community

• Active development with recent updates (V3.x.x) introducing a new UI, optimized node selection, and a plugin system.
• Community support available via QQ Group and Telegram Group.
• Developer support via afdian.

Licensing & Compatibility

• The project is described as "completely free" but does not explicitly state a license in the README.
• Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

• macOS users may need to disable system security checks (sudo spctl --master-disable) or use specific codesign commands to run the application.
• The README mentions future implementation of voice recognition but does not provide a timeline.