The model was presented in the paper Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets.

Recent advances in 3D-native generative models have accelerated asset creation for games, film, and design. However, most methods still rely primarily on image or text conditioning and lack fine-grained, cross-modal controls, which limits controllability and practical adoption. To address this gap, we present Hunyuan3D-Omni, a unified framework for fine-grained, controllable 3D asset generation built on Hunyuan3D 2.1. In addition to images, Hunyuan3D-Omni accepts point clouds, voxels, bounding boxes, and skeletal pose priors as conditioning signals, enabling precise control over geometry, topology, and pose. Instead of separate heads for each modality, our model unifies all signals in a single cross-modal architecture. We train with a progressive, difficulty-aware sampling strategy that selects one control modality per example and biases sampling toward harder signals (e.g., skeletal pose) while downweighting easier ones (e.g., point clouds), encouraging robust multi-modal fusion and graceful handling of missing inputs. Experiments show that these additional controls improve generation accuracy, enable geometry-aware transformations, and increase robustness for production workflows.

Hunyuan3D-Omni is a unified framework for the controllable generation of 3D assets, which inherits the structure of Hunyuan3D 2.1. In contrast, Hunyuan3D-Omni constructs a unified control encoder to introduce additional control signals, including point cloud, voxel, skeleton, and bounding box.

It takes 10 GB VRAM for generation.

We test our model with Python 3.10.

The control_type parameter has four available options:

point: Use point control type for inference.
voxel: Use voxel control type for inference.
bbox: Use bounding box control type for inference.
pose: Use pose control type for inference.

The --use_ema flag enables the use of Exponential Moving Average (EMA) model for more stable inference.

The --flashvdm flag enables FlashVDM optimization for faster inference speed.

Please choose the appropriate control_type based on your requirements. For example, if you want to use the point control type, you can run:

We would like to thank the contributors to the TripoSG, Trellis, DINOv2, Stable Diffusion, FLUX, diffusers, HuggingFace, CraftsMan3D, Michelangelo, Hunyuan-DiT, HunyuanVideo, HunyuanWorld-1.0, and HunyuanWorld-Voyager repositories, for their open research and exploration.

If you use this code in your research, please cite: