A charming AI companion built with modern web technologies. This playful chat application demonstrates real-time streaming responses, rate limiting, and contemporary UI design patterns.
flirtatiousAI creates engaging, witty conversations using Meta's Llama 3.2 model via Hugging Face. The AI personality is designed to be naturally flirtatious with subtle innuendo, playful charm, and a quirky fascination with feet, while maintaining tasteful boundaries and respectful interaction.
Clone the repository
git clone https://github.com/MaDrCloudDev/flirtatiousAI.git
cd flirtatiousAI
Install dependencies
pnpm install
Configure environment
Create a .env
file with your Hugging Face token:
HF_ACCESS_TOKEN=your_hugging_face_token_here
Get your token from Hugging Face Settings (requires fine-grained token with inference.serverless.write
permissions)
Start development server
pnpm dev
The application follows a clean, modern architecture:
The application implements IP-based rate limiting (3 requests per 5 minutes) to prevent abuse and manage API costs. This is handled in-memory and resets on server restart.
This is a personal project showcasing modern web development practices. Feel free to fork and experiment with your own modifications.