NVIDIAGameWorks/kaolin
A PyTorch Library for Accelerating 3D Deep Learning Research
Provides GPU-optimized operations for multi-representation 3D workflows including differentiable mesh rendering (DIB-R), fast format conversions, Structured Point Cloud acceleration structures, and physics simulation with collision detection. Integrates differentiable camera and lighting APIs with spherical harmonics/gaussians, plus native support for 3D gaussian splats and volumetric rendering modes. Built on CUDA kernels and designed for end-to-end differentiable pipelines across neural rendering, geometry optimization, and simulation tasks.
5,056 stars. Actively maintained with 10 commits in the last 30 days.
Stars
5,056
Forks
618
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 13, 2026
Commits (30d)
10
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/NVIDIAGameWorks/kaolin"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
Jim137/qkan
PyTorch implementation of QKAN "Quantum-inspired Kolmogorov-Arnold Network"...
AntonioTepsich/Convolutional-KANs
This project extends the idea of the innovative architecture of Kolmogorov-Arnold Networks (KAN)...
lgy112112/ikan
ikan: many kan variants for every body
Indoxer/LKAN
Variations of Kolmogorov-Arnold Networks
RomanBresson/KAGNN
This is the official repository for our paper KAGNNs: Kolmogorov-Arnold Networks meet Graph Learning.