GLB Compressor

Runs in your browser 200 MB Limit3D Tools#WebXR

Shrink GLB 3D models for the web with Draco or Meshopt, WebP textures and a target-size mode. Runs in your browser.

3D studio · GLB and glTF

Drop a 3D model to inspect and shrink it

Preview it in 3D, see exactly what makes it heavy, then compress it with Draco or Meshopt geometry and WebP textures, or let Smart mode find the gentlest settings that hit your target size.

  • Draco & Meshopt
  • WebP textures
  • Smart target size
  • Runs in your browser

A textured, animated model built on this device. Nothing is downloaded.

Your model never leaves this device: inspection, preview and compression all run in your browser.

How to use, limits & privacy

About GLB Compressor

Inspect, preview and shrink GLB models without uploading them. See what makes a model heavy (geometry, textures, animations), then compress it with Draco or Meshopt, mesh simplification and WebP textures, or let Smart mode find the lightest settings that fit a target size. Compare the original and the result in the 3D viewer before you download.

How to Use

1

Open a model

Drop a .glb file (or a .gltf with embedded data) onto the page. The viewer and the details panel show its meshes, textures, animations and suggestions.

2

Choose Smart or Manual

Smart mode fits a target size with the least quality loss. Manual mode lets you pick a preset or set geometry, simplification and texture options yourself.

3

Compare and download

Switch between Original and Compressed in the viewer, check the size and triangle counts, then download the compressed model.

Privacy & Processing

  • Mode: local
  • Files Leave Browser: Local tool processing; review details below
  • Max Input Size: 200 MB
  • Account Required: No
  • Data Stored Locally: Nothing is stored. Models stay in memory only while this page is open.
  • Network Processing: Assets or models may require an initial download

Models are inspected, previewed and compressed in your browser and never uploaded.

Rules & Limitations

  • Models up to 200 MB. Very large models may be slow or fail on phones.
  • GLB files, or .gltf files with all data embedded. A .gltf that points to separate files must be packed into a GLB first.
  • KTX2 textures are kept as they are.
  • If the result is not smaller, you are offered the original file instead.

Top Suggestions

  • Making product and AR models load faster on websites
  • Getting a model under a marketplace or platform upload limit
  • Finding which textures or meshes make a model heavy
  • Previewing GLB files and their animations without installing anything

GLB Compressor FAQ

How do I compress a GLB file?

Drop the .glb file on this page. Choose Smart and enter the size you need, or choose Manual and pick a preset (Light, Balanced or Aggressive) or your own settings. Compare the original and compressed model in the 3D viewer, then download the smaller GLB.

Is my 3D model uploaded to a server?

No. Inspection, preview and compression all run in your browser, so the model never leaves your device. That also means there is no queue, account or daily limit.

Why is my GLB file so large?

Usually because of textures: a few 4K PNG or JPEG images can outweigh the rest of the model. Dense meshes with many triangles and long animations come next. The Details panel shows what takes the space and suggests what to fix first.

Can I compress a GLB to a specific size?

Yes. Smart mode takes a target in MB and tries progressively stronger settings, from lossless Meshopt cleanup to Draco with WebP textures at 2K or 1K, and then mesh simplification. It keeps the gentlest result that fits, for example a 5 MB limit for a product page or AR viewer.

Draco or Meshopt: which should I use?

Draco usually gives the smallest geometry. Meshopt decodes faster and also compresses animation data. Both need a decoder in the viewer; three.js, Babylon.js and <model-viewer> support both. If your app can't decode them, choose Quantize, which needs no decoder.

Does compression reduce quality?

Draco and Meshopt geometry compression is barely visible at the default settings. Smaller textures, lower WebP quality and mesh simplification trade detail for size, so check the result in the split view, side by side with the original, before you download.

Does it work with glTF files and keep animations?

It opens .glb files and .gltf files with all data embedded, and saves a GLB. A .gltf that points to separate .bin or image files must be packed into a GLB first. Animations are kept, and the preview plays them so you can check.

Will the compressed GLB work in three.js, Unity or Blender?

Yes, it is a standard GLB. three.js, Babylon.js, <model-viewer> and Blender read Draco-compressed GLBs; in Unity, glTFast needs its Draco and Meshopt packages. For apps without a decoder, compress with Quantize or None.