GaiZhenbiao/ChuanhuChatGPT

What it solves

Chuanhu Chat provides a lightweight, feature-rich web graphical user interface (GUI) for interacting with various Large Language Models (LLMs), eliminating the need to rely solely on official web interfaces or complex command-line tools.

How it works

It acts as a frontend wrapper built on Gradio that connects to LLMs via APIs (such as OpenAI, Azure, Google Gemini, Claude, and DeepSeek) or through local deployments (such as ChatGLM, LLaMA, and Qwen). It includes integrated tools for enhancing model capabilities, such as web search and document-based Q&A.

Who it’s for

It is designed for users who want a polished, customizable interface for multiple AI models, developers who need a way to deploy local LLMs quickly, and power users who want advanced features like API load balancing and fine-tuning capabilities.

Highlights

• Broad Model Support: Compatible with major API-based models (GPT-5, GPT-4o, Claude 3.5, DeepSeek R1) and locally hosted models.
• Enhanced Capabilities: Includes an AI assistant (similar to AutoGPT), real-time web search, and a knowledge base for file-based questioning.
• Advanced UI/UX: Features a mobile-responsive design, PWA support, dark/light mode, and a refined "Small-and-Beautiful" theme with glassmorphism effects.
• Power User Tools: Supports GPT-3.5 fine-tuning, multi-API key load balancing, custom proxy settings, and regex-based history search.
• Conversation Management: Offers automatic session naming by the LLM, multi-user history isolation, and prompt templates for role-playing.