Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation

Papers with Code Trending Papers
3d-generation diffusion-model texture-synthesis shape-generation open-source tencent

Summary

Hunyuan3D 2.0 is a scalable flow-based diffusion transformer system for high-resolution textured 3D asset generation, outperforming state-of-the-art models and publicly released with code and weights.

We present Hunyuan3D 2.0, an advanced large-scale 3D synthesis system for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scale shape generation model -- Hunyuan3D-DiT, and a large-scale texture synthesis model -- Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. The texture synthesis model, benefiting from strong geometric and diffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we build Hunyuan3D-Studio -- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2
Original Article
View Cached Full Text

Cached at: 07/12/26, 04:48 PM

Paper page - Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generation

Source: https://huggingface.co/papers/2501.12202 Published on Jan 21, 2025

·

Submitted byhttps://huggingface.co/akhaliq

AKon Jan 22, 2025

Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

Hunyuan3D 2.0, an advanced 3D synthesis system, uses flow-based diffusion transformers for shape generation and diffusion priors for texture synthesis, outperforming existing models in geometry, condition alignment, and texture quality.

We present Hunyuan3D 2.0, an advanced large-scale3D synthesissystem for generating high-resolution textured 3D assets. This system includes two foundation components: a large-scaleshape generationmodel --Hunyuan3D-DiT, and a large-scaletexture synthesismodel --Hunyuan3D-Paint. The shape generative model, built on a scalable flow-based diffusion transformer, aims to create geometry that properly aligns with a given condition image, laying a solid foundation for downstream applications. Thetexture synthesismodel, benefiting from strong geometric anddiffusion priors, produces high-resolution and vibrant texture maps for either generated or hand-crafted meshes. Furthermore, we buildHunyuan3D-Studio-- a versatile, user-friendly production platform that simplifies the re-creation process of 3D assets. It allows both professional and amateur users to manipulate or even animate their meshes efficiently. We systematically evaluate our models, showing that Hunyuan3D 2.0 outperforms previous state-of-the-art models, including the open-source models and closed-source models in geometry details, condition alignment, texture quality, and etc. Hunyuan3D 2.0 is publicly released in order to fill the gaps in the open-source 3D community for large-scale foundation generative models. The code and pre-trained weights of our models are available at: https://github.com/Tencent/Hunyuan3D-2

View arXiv pageView PDFGitHub14.2kautoAdd to collection

Get this paper in your agent:

hf papers read 2501\.12202

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper54

#### tencent/Hunyuan3D-2 Image-to-3D• UpdatedOct 17, 2025 • 94.4k • 1.79k #### tencent/Hunyuan3D-2.1 Image-to-3D• UpdatedOct 17, 2025 • 37.5k • 1.06k #### tencent/Hunyuan3D-2mv Image-to-3D• UpdatedOct 17, 2025 • 3.64k • 416 #### tencent/Hunyuan3D-Omni Image-to-3D• UpdatedOct 17, 2025 • 1.45k • 176 Browse 54 models citing this paper## Datasets citing this paper1

#### tencent/HY3D-Bench UpdatedApr 9 • 6.77k • 93

Spaces citing this paper476

Browse 476 spaces citing this paper## Collections including this paper33

Browse 33 collections that include this paper

Similar Articles

tencent/HY-World-2.0

Hugging Face Models Trending

HY-World 2.0 is Tencent's open-source multi-modal 3D world model that reconstructs and generates 3D worlds from text, images, and videos, producing editable 3D assets (meshes/Gaussian Splatting) comparable to closed-source methods.