Helps you find AI models for your task, prioritizing smaller and more efficient options.
- Describe your AI task in plain language
- Get model recommendations ranked by efficiency (smaller first)
- See environmental impact estimates
Supports: Computer Vision, NLP, Speech, Time Series, Recommendations, Reinforcement Learning, Data Preprocessing
npm install
npm run dev # Start dev server
npm run build # Production build
npm test # Run tests- Runs entirely in the browser (no backend)
- Uses MiniLM embeddings (~23MB, downloaded on first use) to classify your task
- Falls back to keyword matching if embeddings aren't available
- PWA - installable and works offline after first visit
Models are organized into 4 tiers by size and capability:
| Tier | Size | Environmental Score | Typical use |
|---|---|---|---|
| Lightweight | ≤500MB | Low | Edge/mobile/browser |
| Standard | ≤4GB | Medium | Cloud, quantized models |
| Advanced | ≤20GB | High | Dedicated GPU, full LLMs |
| Extra Large | >20GB | High | High-end GPU, 13B+ models |
Automatically deploys to GitHub Pages on push to main.
Manual: npm run build and deploy the dist/ folder.
See docs/ for technical details.
MIT