The 3D Model Compressor is now live at /compress. Reduce your 3D file sizes by 30–70% without visible quality loss — entirely in your browser.
Three Compression Presets
Pick the level that fits your use case:
- Light (10–30% reduction) — Almost no quality loss, ideal for precision engineering models
- Balanced (30–50% reduction) — Best trade-off between file size and visual fidelity
- Maximum (50–70% reduction) — Aggressive compression for web, AR, and mobile delivery
Side-by-Side Comparison
After compression, an interactive slider lets you compare the original and compressed models. Drag the divider to reveal differences — or rather, see that there are none.
On desktop, both models render simultaneously with synchronized camera controls. On mobile, a tab switcher keeps things fast with a single WebGL context.
Compression Stats
Every compression shows detailed statistics:
- Original vs compressed file size
- Percentage saved
- Vertex and face counts (unchanged — geometry is preserved)
- Compression time
How It Works
The compressor uses glTF-Transform to apply four optimizations:
- Dedup — Remove duplicate geometry, materials, and textures
- Weld — Merge vertices that share the same position
- Prune — Strip unused nodes, materials, and accessors
- Quantize — Reduce vertex precision from 32-bit to 10–14 bit (imperceptible difference)
All processing runs locally in your browser. Your files are never uploaded anywhere.
SEO Format Pages
Each compressible format now has its own page (e.g., /compress/glb) with targeted metadata and JSON-LD structured data.