Image-to-3D
Hunyuan3D-2
English
Chinese
nielsr HF Staff commited on
Commit
8df2582
·
verified ·
1 Parent(s): 8bb751a

Refactor metadata to YAML front matter and add GitHub badge

Browse files

This PR improves the model card for Hunyuan3D-Omni by:

* **Refactoring Metadata**: Moved the existing metadata from an inline YAML block within the Markdown content to the proper YAML front matter at the top of the `README.md`. This ensures the Hub correctly processes the metadata.
* **Adding Paper Information**: Included the paper title and abstract at the beginning of the model card for immediate context and improved discoverability.
* **Adding GitHub Badge**: A dedicated badge linking to the official GitHub repository (`https://github.com/Tencent-Hunyuan/Hunyuan3D-Omni`) has been added to the top section, making the code more accessible.

Other existing information, including the `library_name`, `license` details, pipeline tags, and usage instructions, remains unchanged as they are already correctly provided or no definitive code evidence was found to warrant a modification to the `library_name` for automated widget compatibility. The existing arXiv paper link is also preserved as per instructions.

Files changed (1) hide show
  1. README.md +15 -7
README.md CHANGED
@@ -1,18 +1,25 @@
1
  ---
 
 
 
2
  library_name: hunyuan3d-2
3
  license: other
4
  license_name: tencent-hunyuan-community
5
  license_link: https://github.com/Tencent-Hunyuan/Hunyuan3D-Omni/blob/main/LICENSE
6
- language:
7
- - en
8
- - zh
9
- tags:
10
- - image-to-3d
11
- - text-to-3d
12
  pipeline_tag: image-to-3d
 
 
 
13
  extra_gated_eu_disallowed: true
14
  ---
15
 
 
 
 
 
 
 
 
16
  <p align="center">
17
  <img src="assets/omni_teaser.png">
18
  </p>
@@ -21,8 +28,9 @@ extra_gated_eu_disallowed: true
21
  <a href=https://3d.hunyuan.tencent.com target="_blank"><img src=https://img.shields.io/badge/Official%20Site-333399.svg?logo=homepage height=22px></a>
22
  <a href=https://huggingface.co/tencent/Hunyuan3D-Omni target="_blank"><img src=https://img.shields.io/badge/%F0%9F%A4%97%20Models-d96902.svg height=22px></a>
23
  <a href=https://3d-models.hunyuan.tencent.com/ target="_blank"><img src= https://img.shields.io/badge/Page-bb8a2e.svg?logo=github height=22px></a>
24
- <a href=https://discord.gg/dNBrdrGGMa target="_blank"><img src= https://img.shields.io/badge/Discord-white.svg?logo=discord height=22px></a>
25
  <a href=https://arxiv.org/pdf/2509.21245 target="_blank"><img src=https://img.shields.io/badge/Report-b5212f.svg?logo=arxiv height=22px></a>
 
 
26
  <a href=https://x.com/TencentHunyuan target="_blank"><img src=https://img.shields.io/badge/Hunyuan-black.svg?logo=x height=22px></a>
27
  <a href="#community-resources" target="_blank"><img src=https://img.shields.io/badge/Community-lavender.svg?logo=homeassistantcommunitystore height=22px></a>
28
  </div>
 
1
  ---
2
+ language:
3
+ - en
4
+ - zh
5
  library_name: hunyuan3d-2
6
  license: other
7
  license_name: tencent-hunyuan-community
8
  license_link: https://github.com/Tencent-Hunyuan/Hunyuan3D-Omni/blob/main/LICENSE
 
 
 
 
 
 
9
  pipeline_tag: image-to-3d
10
+ tags:
11
+ - image-to-3d
12
+ - text-to-3d
13
  extra_gated_eu_disallowed: true
14
  ---
15
 
16
+ # Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets
17
+
18
+ The model was presented in the paper [Hunyuan3D-Omni: A Unified Framework for Controllable Generation of 3D Assets](https://huggingface.co/papers/2509.21245).
19
+
20
+ ## Abstract
21
+ 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.
22
+
23
  <p align="center">
24
  <img src="assets/omni_teaser.png">
25
  </p>
 
28
  <a href=https://3d.hunyuan.tencent.com target="_blank"><img src=https://img.shields.io/badge/Official%20Site-333399.svg?logo=homepage height=22px></a>
29
  <a href=https://huggingface.co/tencent/Hunyuan3D-Omni target="_blank"><img src=https://img.shields.io/badge/%F0%9F%A4%97%20Models-d96902.svg height=22px></a>
30
  <a href=https://3d-models.hunyuan.tencent.com/ target="_blank"><img src= https://img.shields.io/badge/Page-bb8a2e.svg?logo=github height=22px></a>
 
31
  <a href=https://arxiv.org/pdf/2509.21245 target="_blank"><img src=https://img.shields.io/badge/Report-b5212f.svg?logo=arxiv height=22px></a>
32
+ <a href=https://github.com/Tencent-Hunyuan/Hunyuan3D-Omni target="_blank"><img src=https://img.shields.io/badge/GitHub-Code-181717.svg?logo=github height=22px></a>
33
+ <a href=https://discord.gg/dNBrdrGGMa target="_blank"><img src= https://img.shields.io/badge/Discord-white.svg?logo=discord height=22px></a>
34
  <a href=https://x.com/TencentHunyuan target="_blank"><img src=https://img.shields.io/badge/Hunyuan-black.svg?logo=x height=22px></a>
35
  <a href="#community-resources" target="_blank"><img src=https://img.shields.io/badge/Community-lavender.svg?logo=homeassistantcommunitystore height=22px></a>
36
  </div>