Chuanhu ChatGPT provides a feature-rich, user-friendly web GUI for interacting with various Large Language Models (LLMs), including ChatGPT, Gemini, and locally deployed models. It targets users who want a polished interface for advanced LLM functionalities like agent-based task completion, file-based Q&A, web search integration, and model fine-tuning.

How It Works

The project leverages Gradio for its UI framework, but with significant custom enhancements to achieve a "small-and-beautiful" aesthetic, including non-linear animations and a glassmorphism effect, distinguishing it from standard Gradio applications. It supports a wide array of LLMs through API calls and local deployments, offering features like system prompts, conversation history management with search and auto-naming, and customizable model parameters.

Quick Start & Requirements

• Install via pip install -r requirements.txt after cloning the repository.
• Requires Python and an API key for services like OpenAI.
• Setup involves cloning, installing dependencies, and configuring config.json.
• Official wiki for detailed installation and usage: https://github.com/GaiZhenbiao/ChuanhuChatGPT/wiki

Highlighted Details

• Supports ChatGPT (GPT-4, GPT-4o), Gemini Pro, Claude 3, and various open-source models (LLaMA, ChatGLM, DeepSeek).
• Features include agent assistants, file-based Q&A, web search, and GPT-3.5 fine-tuning.
• Enhanced UI with PWA installation support, mobile adaptation, and auto/dark mode switching.
• Advanced options for custom API hosts, proxies, and API key load balancing.

Maintenance & Community

The project is actively maintained by GaiZhenbiao. Community engagement is encouraged via Telegram groups.

Licensing & Compatibility

The repository does not explicitly state a license in the provided README. Users should verify licensing for commercial use or integration with closed-source projects.

Limitations & Caveats

Some models like XMChat and Midjourney do not support streaming. The project's reliance on Gradio, despite customizations, might still impose certain UI/UX constraints. Users encountering issues are advised to consult the extensive FAQ page.