Home▸AI & Machine Learning▸Model Compression Lab — Shrink a Network Without Wrecking It

💾 Model Compression Lab — Shrink a Network Without Wrecking It

Apply pruning and quantization to a neural network and watch model size and accuracy trade off against each other in real time.

AI & Machine Learning3DAdvanced60 FPS📱 Mobile-adapted

⚙ Under the hood

Apply pruning and quantization to a neural network and watch model size and accuracy trade off against each other in real time.

neural networksmodel compressionpruningquantizationmachine learningdeep learningThree.js

3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install