SEFurkan GözükaraSECourses · Generative AI Join onPatreon

Hunyuan3D-1 Released: Open Source Text-to-3D and Image-to-3D Model

Hunyuan3D-1 by Tencent is a SOTA open source model for text-to-3D and image-to-3D that makes multi-view images in about 4 seconds and a mesh in about 7.

Announcement of the 1-click installers to install and use it both locally on Windows and on the RunPod and Massed Compute cloud; it works well on 24 GB GPUs. Explains the two-stage approach with a lite and a standard version that both support text and image input.

Updated By Dr. Furkan Gözükara
WindowsRunPodMassed Compute1-Click InstallerNews
Hunyuan3D-1 Released: Open Source Text-to-3D and Image-to-3D Model
On this page
Older post. this is an older post kept for reference. The current, maintained version is TRELLIS 2 App: Image to 3D Mesh & Texture With UniRig Auto Rigging.

Hunyuan3D-1 - SOTA Open Source Text-to-3D and Image-to-3D - 1-Click Install and use both Locally on Windows and on Cloud - RunPod and Massed Compute

Installers are here : Hunyuan3D-1 1-Click Installers: Text to 3D & Image to 3D Generation

Automatic Installers

So what is Hunyuan3D-1

Official repo : https://github.com/tencent/Hunyuan3D-1

Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation

Abstract

So what is Hunyuan3D-1 – Hunyuan3D-1 Released: Open Source Text-to-3D and Image-to-3D Model

While 3D generative models have greatly improved artists' workflows, the existing diffusion models for 3D generation suffer from slow generation and poor generalization. To address this issue, we propose a two-stage approach named Hunyuan3D-1.0 including a lite version and a standard version, that both support text- and image-conditioned generation.

In the first stage, we employ a multi-view diffusion model that efficiently generates multi-view RGB in approximately 4 seconds. These multi-view images capture rich details of the 3D asset from different viewpoints, relaxing the tasks from single-view to multi-view reconstruction. In the second stage, we introduce a feed-forward reconstruction model that rapidly and faithfully reconstructs the 3D asset given the generated multi-view images in approximately 7 seconds. The reconstruction network learns to handle noises and in-consistency introduced by the multi-view diffusion and leverages the available information from the condition image to efficiently recover the 3D structure.

Our framework involves the text-to-image model, i.e., Hunyuan-DiT, making it a unified framework to support both text- and image-conditioned 3D generation. Our standard version has 3x more parameters than our lite and other existing model. Our Hunyuan3D-1.0 achieves an impressive balance between speed and quality, significantly reducing generation time while maintaining the quality and diversity of the produced assets.

↑ ↓ to move · Enter to open · Esc to close · Ctrl K or / to search from any page