For those who don’t know the tool: glTF sits between DCC export and runtime playback, and these web tools help you sanity check assets and rendering across engines before shots get spicy.

Two sites, one mission: fewer surprises

The glTF Sample Assets site and the glTF Render Fidelity Comparison site just got meaningful upgrades. They target the same everyday pain: you export a model, it looks right in one viewer, and then it looks slightly off somewhere else where it matters. The refreshed Sample Assets experience focuses on browsing, filtering, and sharing reference models that exercise specific features of glTF. The updated Render Fidelity Comparison experience focuses on visual baselines, with side-by-side checks that make renderer differences harder to hand wave and easier to track down. If you build tools, ship engines, validate exporters, or wrangle assets for realtime and VFX, these sites are designed to be the boring, reliable kind of helpful.

Sample Assets gets a real front door

The new landing experience for glTF Sample Assets presents the repository as a proper website. The intent is simple: make it easier to browse and find models by what they demonstrate. Models appear with thumbnails, and animated assets include a small animation preview. The site supports light and dark themes and aims for fast loading and fluid navigation. Filtering and search are first class, with tags that map to features, functionality, and extension coverage.

When you open a model, you land on a dedicated model page with an interactive preview, metadata, and download links. The preview uses the glTF Sample Renderer, which is built on WebGL and targets correct support for official KHR extensions. A context menu in the preview provides presentation options that include debug visualization modes for material properties, variant selection when multiple variants exist, and wireframe presentation when the WEBGL polygon mode extension is available.

The model page also links out to external tools and shows a recommendation style list of related assets, which is a surprisingly effective way to bounce between test cases when you are chasing a specific shading edge case across multiple files.

One practical detail matters here: each model includes a screenshot, description, license information, and a live preview. That combination turns the site into a quick reference shelf rather than a scavenger hunt through folders.

Render Fidelity Comparison leans into side by side truth

The redesigned glTF Render Fidelity Comparison site takes the same library mindset and aims it at output consistency. It frames the core goal as convergence around physically based rendering materials so you can expect a model to appear as intended across renderers and lighting environments. It also notes that this is ambitious, with realtime rendering quality still changing rapidly. *cough* understatement *cough*

The workflow is built for quick comparisons. You choose a model, filter by tags or search, then pick two renders and select a view mode. The comparison modes include side by side viewing, a slider view, and a difference view that shows computed differences.

The site spans popular realtime web renderers and also includes path tracers. In practice, that means you can jump between quick interactive engines like babylon.js and three.js, then also look at path traced production renderings from Blender using Cycles and from V-Ray. It is not about crowning a winner. It is about seeing where two implementations diverge so you can reproduce, triage, and fix.

Comparison images are generated offline and submitted via pull requests to the site repository, which keeps the content auditable and repeatable. That approach also makes it easier for teams to add coverage for their own renderer outputs without inventing a new pipeline.

The biggest new trick is an experimental 3D comparison mode that enables full interactivity comparison of two realtime renderers side by side. It currently supports three.js, babylon.js, model-viewer, and three js path tracer tooling. If you have ever tried to discuss a subtle lighting mismatch over screenshots, you know why interactive side by side matters. In one sentence, the updated site encourages you to comapre outputs with fewer assumptions and more pixels on screen.

https://github.khronos.org/glTF-Render-Fidelity/model/AttenuationTest

Contributing is part of the design

Both sites are open to contributions. For sample assets, new models can be proposed via pull requests to the repository. For render fidelity, you can submit renderer outputs by following the repository guidelines for generating renders of a sample asset and placing the results in the appropriate location for comparison.

There is also a straightforward feedback loop via issues in the repositories, which keeps bug reports and feature requests attached to something actionable. This is the one place where it is worth naming the organizer exactly once: Khronos is clearly aiming to make the glTF developer ecosystem easier to navigate, easier to test, and harder to break quietly.

https://github.khronos.org/glTF-Assets/

https://github.khronos.org/glTF-Render-Fidelity/

The post glTF sample assets and fidelity site refresh first appeared on DIGITAL PRODUCTION and was written by Bela Beier.

Khronos Group, in collaboration with OGC, Niantic Spatial, Cesium (Bentley), and Esri, has formally added 3D Gaussian Splats (3DGS) to the glTF ecosystem. Two experimental extensions, KHR_gaussian_splatting and KHR_gaussian_splatting_compression_spz, lay the foundation for interoperable splat workflows across DCCs, render engines, geospatial platforms, and the web.

This development places Gaussian Splats, until now a fast-moving research and tool-integration phenomenon, onto the same structural footing as meshes, point clouds, and textures. In other words: splats just got standardized.

From Research Trick to File Format

Gaussian splatting isn’t new. The math dates back to the 1990s, but the 2023 Inria paper “3D Gaussian Splatting for Real-Time Radiance Field Rendering” gave the technique a second life. By rendering 3D scenes as clouds of anisotropic Gaussian ellipsoids, each storing radiance and opacity, researchers achieved real-time neural rendering that rivaled neural radiance fields (NeRFs) without the training overhead.

What makes splats compelling is their efficiency: no polygons, no voxel grids, no heavy neural nets. Just point-based primitives that can capture detail in thin structures, soft translucency, or reflective materials.

Within a year, Houdini gained a native GSOPs plugin, V-Ray added splat rendering in Chaos Vantage, and studios began experimenting with splats in volumetric video pipelines such as HOLOSYS+. Now, the Khronos Group has moved to lock this fast-growing format into glTF, which already brands itself as the “JPEG of 3D.”

KHR_gaussian_splatting

The first of the new extensions, KHR_gaussian_splatting, defines how splats are represented inside glTF assets. Each splat is treated as a point primitive with the following attributes: Position (XYZ), Rotation (quaternion), Scale (anisotropic ellipsoid dimensions), Opacity and Spherical Harmonics coefficients for both diffuse and specular lighting

The extension allows fallback to sparse point clouds if a consuming renderer doesn’t support splats—ensuring that files remain viewable, even without specialized splat engines. For shader developers, the inclusion of spherical harmonics (SH) is key. SH enables compact representation of lighting across a hemisphere, which allows splats not only to look correct but also to respond naturally to lighting in real-time environments. Up to 3rd-degree SH coefficients are supported.

KHR_gaussian_splatting_compression_spz

The second extension, KHR_gaussian_splatting_compression_spz, tackles the elephant in the room: storage size. Traditional PLY splat datasets are enormous, with billions of points often required to reconstruct complex scenes. To address this, Niantic Spatial contributed the SPZ compression format, MIT-licensed for free use. SPZ achieves up to 90% smaller file sizes compared to equivalent PLY datasets while maintaining visual fidelity.

SPZ blobs can be embedded directly into glTF buffers. At runtime, they can either be streamed as compressed chunks or decompressed into raw primitives. This dual mode makes SPZ especially suited for geospatial and AR contexts where bandwidth is limited.

SPZ v2.0: Better Rotations

Version 2.0.0 of SPZ adds improved encoding of splat orientations. Rotations are stored as quaternions, but instead of four floats, SPZ encodes only the three smallest components as 10-bit signed integers. The missing largest component is reconstructed at runtime, with a 2-bit index flagging which component was omitted.

This compact scheme increases precision for long, thin geometry—think power lines, fences, antennas—and improves rendering stability in massive geospatial datasets. It’s the difference between “the wire looks straight” and “the wire jitters like a bad motion-capture solve.”

Why This Matters

1. Standardization
   Until now, Gaussian Splatting workflows were fragmented: PLY exports here, custom Houdini nodes there, and bespoke converters everywhere. With glTF extensions, splats can flow through established pipelines with predictable interoperability.
2. Production Viability
   Splats are great in research demos, but production requires stability. glTF’s adoption across web, AR, and DCC ecosystems ensures that Gaussian Splats won’t remain academic curiosities.
3. Performance & Compression
   SPZ addresses the practical bottleneck: dataset size. Geospatial datasets and volumetric captures can hit terabytes in raw splats; with SPZ, they become streamable assets.
4. Lighting Flexibility
   By embedding spherical harmonics, splats can sit naturally in lighting pipelines—whether inside Houdini’s Karma XPU, V-Ray, or CesiumJS streaming environments.

glTF as the “JPEG of 3D”

glTF’s trajectory has always been about unification and accessibility. From PBR textures to Draco compression, glTF brought complex 3D assets into a web- and engine-friendly format. Adding Gaussian Splats continues that trend, ensuring that cutting-edge formats don’t remain vendor-locked.

Roadmap and Next Steps

The current extensions are experimental—Khronos invites developers to test, validate, and propose refinements. Early validation across diverse geospatial datasets confirmed SPZ’s effectiveness, but the long-term plan is to extend the extensions with more attributes, compression modes, and streaming optimizations.

The post 3D Gaussian Splats Officially Added to glTF Standard first appeared on DIGITAL PRODUCTION and was written by Bela Beier.