ASCII Art Generator

Inspiration

The inspiration came from wanting to democratize access to AI models and showcase their creative potential. We noticed that many developers struggle to discover and experiment with free AI models, so we created a tool that not only lists available models but also demonstrates their capabilities through the fun medium of ASCII art generation.

What it does

This TypeScript utility automatically discovers all free models available on GitHub Models, generates creative prompts using random model selection, and creates ASCII art through AI-powered generation. It features intelligent retry logic for handling API timeouts, automatic model switching when errors occur, and comprehensive documentation updates with generated art and model attribution.

How we built it

We built this project using TypeScript for type safety and Node.js for the runtime environment. The architecture consists of three main modules: a model discovery service that fetches available models from GitHub's catalog, an API client that handles communication with GitHub Models, and an ASCII art generator that orchestrates the entire process. We implemented robust error handling with timeout management and automatic retry mechanisms.

Challenges we ran into

The biggest challenges were implementing reliable timeout handling across different AI models, managing varying response formats from different models, and ensuring proper ASCII art formatting with bracket closure. We also had to handle rate limits effectively and create seamless fallback mechanisms when specific models became unavailable or unresponsive.

Accomplishments that we're proud of

We're proud of creating a robust system that gracefully handles API failures and automatically switches between models. The random model selection feature makes each generation unique and showcases the diversity of available AI models. We also successfully implemented comprehensive documentation that updates automatically with each generation, creating a living showcase of the tool's capabilities.

What we learned

We gained deep insights into GitHub Models API integration and learned effective patterns for handling distributed AI services. The project taught us the importance of retry mechanisms in AI applications and how to create user-friendly interfaces for complex AI workflows. We also discovered how different models have unique characteristics and response patterns.

What's next for OpenRouter Free Models & ASCII Art Generator

We plan to expand support to include image generation models, implement model performance analytics to help users choose the best models for their needs, create a web interface for easier interaction, add batch processing capabilities, and integrate with additional AI model providers beyond GitHub Models.

Built with

• TypeScript
• Node.js
• GitHub Models API
• dotenv
• tsx
• pnpm
• Fetch API

Built With

• kiro