Favors quality
Keeps the model closest to the original. Safe for inspection and review.
Shrink a model without leaving the browser. Presets trade quality against size; the viewport stays put while you compare them.
Source is auto-detected; the workbench opens directly on the compress task.
or drop model files here
Keeps the model closest to the original. Safe for inspection and review.
Trades a little quality for a meaningfully smaller file.
Smallest result. Best for web delivery where detail matters least.
GLB compresses well because the pipeline can use Draco and texture optimisation; STL has almost nothing to strip.
Open the format-aware compression workflow for OBJ models.
OpenSTLOpen the format-aware compression workflow for STL models.
OpenFBXOpen the format-aware compression workflow for FBX models.
Open3DSOpen the format-aware compression workflow for 3DS models.
OpenDAEOpen the format-aware compression workflow for DAE models.
OpenPLYOpen the format-aware compression workflow for PLY models.
Open3MFOpen the format-aware compression workflow for 3MF models.
OpenAMFOpen the format-aware compression workflow for AMF models.
OpenOFFOpen the format-aware compression workflow for OFF models.
OpenDXFOpen the format-aware compression workflow for DXF models.
OpenVRMLOpen the format-aware compression workflow for VRML models.
OpenX3DOpen the format-aware compression workflow for X3D models.
OpenXOpen the format-aware compression workflow for X models.
Open3DMOpen the format-aware compression workflow for 3DM models.
OpenGLBOpen the format-aware compression workflow for GLB models.
OpenglTFOpen the format-aware compression workflow for glTF models.
OpenSTEPOpen the format-aware compression workflow for STEP models.
OpenIGESOpen the format-aware compression workflow for IGES models.
OpenUSDZOpen the format-aware compression workflow for USDZ models.
OpenVTKOpen the format-aware compression workflow for VTK models.
OpenXYZOpen the format-aware compression workflow for XYZ models.
OpenPTSOpen the format-aware compression workflow for PTS models.
OpenUSDAOpen the format-aware compression workflow for USDA models.
OpenDrag and drop your 3D model. The browser prepares it for compression.
Choose a quality intent for geometry detail and target size. Compression starts only after you confirm the tradeoff.
Preview validated candidates side-by-side with the original. If no candidate is safely smaller, FabRapid keeps the original file.
FabRapid separates geometry encoding, lossy simplification, and format rewriting so each result can be validated against the correct quality and compatibility rules.
Geometry simplification reduces the number of triangles in the mesh while preserving the overall shape. The algorithm merges nearby vertices and removes triangles that contribute least to the visual appearance, controlled by a simplification ratio you can adjust.
For native GLB files, FabRapid selects Meshopt or Draco from the model content and quality intent. The candidate must pass codec read-back, semantic, Viewer, resource, and whole-file benefit gates.
The current GLB strategy preserves embedded texture bytes. Texture recompression will be added only after separate color, alpha, data-texture, quality, memory, and compatibility validation.
Savings depend on the model content and its existing encoding. FabRapid publishes a candidate only when it is safely smaller than the whole original file; there is no fixed reduction guarantee.
Each quality intent has a different geometry budget. Candidates must pass semantic checks and the product Viewer before download, and the original is kept when those gates fail.
Native GLB uses the new format-aware Meshopt or Draco strategy. Other listed formats currently use the legacy round trip and are delivered only when the final same-format file is safely smaller.
Files above the product limit are rejected before processing. Smaller files can also be rejected when the estimated decoded geometry and codec working set exceed the browser resource budget.
Currently, files are compressed one at a time. Upload your file, adjust settings, download the compressed result, then process the next file.