HugAi is an open-source, full-stack AI assistant built with Spring Boot and Vue.js, designed for users who want to deploy their own AI chat and image generation platform. It integrates multiple large language models and AI art platforms, offering features like streaming responses, role-playing, and cloud storage for conversations.

How It Works

The system utilizes a Java Spring Boot backend for core logic and API integrations, with a Vue.js frontend for the user interface. It supports both SSE and WebSocket for streaming responses, allowing for a dynamic, typewriter-like output. Key architectural choices include a configurable API key pool, support for various cloud storage solutions (MinIO, Qiniu Cloud), and robust authentication via Spring Security for multi-user roles (admin, user, guest).

Quick Start & Requirements

• Installation: Backend (Java 17, Spring Boot 2.7+, Maven, MySQL 8.0, Redis, RabbitMQ) and Frontend (Vue 2, Element UI). Deployment instructions are in the Help Documentation.
• Prerequisites: Java 17+, MySQL 8.0, Redis, RabbitMQ, Node.js for frontend.
• Setup Time: Estimated to be several hours depending on familiarity with the stack.

Highlighted Details

• Supports multiple LLMs: OpenAI (GPT-3.5, GPT-4, GPT-3.5-16K), Xunfei Spark, Baidu Wenxin Yiyan, Ollama.
• Integrates AI art generation: Midjourney, Stable Diffusion, DALL-E with features like text-to-image, image-to-image, upscaling, and prompt optimization.
• Offers role-playing capabilities for various content creation tasks (e.g., social media copy, novel writing).
• Features a configuration center for managing models, API keys, and image generation parameters.

Maintenance & Community

The project is actively maintained by the original author. A QQ discussion group (747193904) is available for community interaction and support.

Licensing & Compatibility

• License: Apache-2.0.
• Commercial Use: Permitted, but requires attribution to the original author for derivative or commercial works.

Limitations & Caveats

The project is under active development, with features like PDF/Word document interaction (vector database integration) and advanced Stable Diffusion integration still in research or early stages. API key management for user-provided keys requires further security enhancements.