News: the 10k dataset is ready for download.

Simulation platform for general-purpose robotics & embodied AI learning.

autosim

✨ An advanced 3D Gaussian Splatting renderer for THREE.js

Code for "EgoX: Egocentric Video Generation from a Single Exocentric Video"

[NeurIPS 2025] InstaInpaint: Instant 3D-Scene Inpainting with Masked Large Reconstruction Model

Full python interactive 3D Gaussian Splatting viewer for real-time editing and analyzing.

Cosmos-Transfer2.5, built on top of Cosmos-Predict2.5, produces high-quality world simulations conditioned on multiple spatial control inputs.

Cosmos-Predict2.5, the latest version of the Cosmos World Foundation Models (WFMs) family, specialized for simulating and predicting the future state of the world in the form of video.

Code for "FlashWorld: High-quality 3D Scene Generation within Seconds" (ICLR 2026 Oral)

[ICLR 2026] Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation

WorldGrow: Generating Infinite 3D World [AAAI 2026 Oral]

Post-training scripts and samples for NVIDIA Cosmos ecosystem

[WIP] Resources for AI engineers. Also contains supporting materials for the book AI Engineering (Chip Huyen, 2025)

[SIGGRAPH Asia 2025] WorldExplorer: Towards Generating Fully Navigable 3D Scenes

[NeurIPS 2025 (Spotlight)] The implementation for the paper "4DGT Learning a 4D Gaussian Transformer Using Real-World Monocular Videos"

3D Gaussian Flats: Hybrid 2D/3D Photometric Scene Reconstruction

Official Implementation of PractiLight: Practical Light Control Using Foundational Diffusion Models

Official Repository of "ROSE: Remove Objects with Side Effects in Videos"

[T-PAMI 2025] DiffMVS & CasDiffMVS

[CVPR'25 Oral] Official implementation for "DiffusionRenderer: Neural Inverse and Forward Rendering with Video Diffusion Models"

Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

[CVPR'25 Oral] MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision

Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels with Hunyuan3D World Model

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models

Advanced viewer implementing several rendering methods for 3D gaussians using Vulkan API

[ICCV 2025 Highlight] Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction