Tencent Hunyuan 3D-2mini Open-source Model - Efficiently Convert Images to 3D with Lightweight Design, Supporting Both Chinese and English Input

Hunyuan3d 2mini

Developed by tencent

Tencent Hunyuan3D 2mini is a lightweight and efficient image-to-3D model with 600 million parameters, supporting both Chinese and English inputs.

3D Vision Supports Multiple LanguagesOpen Source License:Other #Image-to-3D Generation #Lightweight 600M Parameters #High-Resolution Textures

Downloads 23.22k

Release Time : 3/12/2025

Model Overview

Hunyuan3D 2mini is a diffusion model developed by Tencent, focusing on generating high-resolution textured 3D assets from images. Compared to the previous 1.1 billion parameter version, this version is more lightweight and efficient.

Model Features

Lightweight and Efficient

600M parameter shape generator, more lightweight than the previous 1.1B parameter version

High-Resolution Output

Supports generating 3D assets with high-resolution textures

Multimodal Input

Supports both images and text as input for 3D content generation

Model Capabilities

Image-to-3D

Text-to-3D

3D Asset Generation

Use Cases

3D Creation

Rapid 3D Model Generation

Quickly generate corresponding 3D models from 2D images

Produces 3D assets usable in scenarios like gaming and film production

Digital Content Creation

Virtual Scene Construction

Generate 3D scene elements from text descriptions

Accelerates the virtual scene construction process

🚀 Hunyuan3D-2

Hunyuan3D-2 is a model that enables the generation of high - resolution textured 3D assets. It supports both image - to - 3D and text - to - 3D conversion, bringing everyone's imagination of creating and manipulating 3D assets to life.

“ Living out everyone’s imagination on creating and manipulating 3D assets.”

🚀 Quick Start

This repository contains the models of the paper Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation.

Hunyuan3D-2mini contains a 0.6B shape generator, which is smaller and faster than the previous 1.1B one.

💻 Usage Examples

Basic Usage

Here is a simple usage:

from hy3dgen.shapegen import Hunyuan3DDiTFlowMatchingPipeline
pipeline = Hunyuan3DDiTFlowMatchingPipeline.from_pretrained(
    'tencent/Hunyuan3D-2mini',
    subfolder='hunyuan3d-dit-v2-mini',
    use_safetensors=True,
    device='cuda'
)
mesh = pipeline(
    image=image,
    num_inference_steps=30,
    octree_resolution=380,
    num_chunks=20000,
    generator=torch.manual_seed(12345),
    output_type='trimesh'
)[0]

For code and more details on how to use it, refer to the Github repository.

🔗 BibTeX

If you found this repository helpful, please cite our report:

@misc{hunyuan3d22025tencent,
    title={Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation},
    author={Tencent Hunyuan3D Team},
    year={2025},
    eprint={2501.12202},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

@misc{yang2024tencent,
    title={Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation},
    author={Tencent Hunyuan3D Team},
    year={2024},
    eprint={2411.02293},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

✨ Community Resources

Thanks for the contributions of community members, here we have these great extensions of Hunyuan3D 2.0:

📄 License

This project is under the Tencent Hunyuan Community License.

🙏 Acknowledgements

We would like to thank the contributors to the DINOv2, Stable Diffusion, FLUX, diffusers and HuggingFace repositories, for their open research and exploration.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご