Instant neural graphics primitives: lightning fast NeRF and more

CUDA accelerated rasterization of gaussian splatting

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

[CVPR 2023 Highlight] Neural Kernel Surface Reconstruction

Differentiable gaussian rasterization with depth, alpha, normal map and extra per-Gaussian attributes, also support camera pose gradient

Code for SpotLessSplats: Ignoring Distractors in 3D Gaussian Splatting built on gsplat codebase.

A modular differential gaussian rasterization library.

Instant NeRF for depth/3d shape estimation of transparent objects