Rebuilds an object from one reference image as procedural, animation-ready Three.js TypeScript code, not a mesh file.
YanWenKun/Hunyuan3D-2-WinPortable
A Windows portable package that bundles the Hunyuan3D 2.0/2.1 image-to-3D Gradio web app so it runs without manual setup.
- Stars
- 1,357
- Forks
- 129
- Open issues
- 52
- Last commit
- 11 months ago
What is it?
This project packages Tencent's Hunyuan3D 2.0 and 2.1 Gradio web apps into a ready-to-run Windows bundle. Users extract an archive, run a launcher, and open a local browser page to generate 3D meshes from images or text, with optional texture generation.
The pack integrates mmgp from the Hunyuan3D-2GP project, which applies VRAM offloading and optimization so the models can run on consumer NVIDIA GPUs. Separate folders provide the 2.0 model with mmgp optimizations, the official 2.0 code without them, the 2.1 model, and API modes. Model files download on demand at startup.
The package targets NVIDIA GPUs with CUDA 12.9 or 12.6 builds and requires a recent driver. Texture generation additionally needs the CUDA Toolkit and Visual Studio Build Tools 2022 installed on the system.
Key features
- 01Runs the Hunyuan3D 2.0 and 2.1 Gradio web apps on Windows without manual environment setup.
- 02Supports image-to-3D mesh generation and optional texture generation.
- 03Includes mmgp VRAM optimizations so models can run on consumer NVIDIA GPUs.
- 04Offers separate builds for CUDA 12.9 and CUDA 12.6 to match GPU architectures.
- 05Downloads model files on demand and can be updated with UPDATE.bat.
Use cases
- Generating 3D meshes from reference images on a local Windows PC with an NVIDIA GPU.
- Trying Hunyuan3D 2.0 or 2.1 without building the Python environment from source.
- Running Hunyuan3D on GPUs with limited VRAM using mmgp offloading.
Topics
FAQ
- What is Hunyuan3D-2-WinPortable?
- A Windows portable package that bundles the Hunyuan3D 2.0/2.1 image-to-3D Gradio web app so it runs without manual setup. This project packages Tencent's Hunyuan3D 2.0 and 2.1 Gradio web apps into a ready-to-run Windows bundle. Users extract an archive, run a launcher, and open a local browser page to generate 3D meshes from images or text, with optional texture generation.
- Is Hunyuan3D-2-WinPortable free to use? What license does it have?
- Hunyuan3D-2-WinPortable is open source under the GPL-3.0 license. Check the license terms before using it in commercial products.
- Is Hunyuan3D-2-WinPortable still maintained?
- No commits for 6+ months. The last commit to Hunyuan3D-2-WinPortable was on Nov 19, 2025.
- What can I use Hunyuan3D-2-WinPortable for?
- Generating 3D meshes from reference images on a local Windows PC with an NVIDIA GPU. Trying Hunyuan3D 2.0 or 2.1 without building the Python environment from source. Running Hunyuan3D on GPUs with limited VRAM using mmgp offloading.
- Where can I get Hunyuan3D-2-WinPortable?
- The source code and installation instructions are on GitHub: https://github.com/YanWenKun/Hunyuan3D-2-WinPortable
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Open-source Stable Diffusion-based tool for generating character-consistent comic images and long-range image-to-video sequences.
An open-source Diffusion model engine for image, video, and audio generation, with low-VRAM inference and training support.
An all-in-one training suite for finetuning diffusion image, video and audio models on consumer GPUs, via a web UI or CLI.
Open-source toolkit for optimizing and deploying deep learning models for inference on Intel CPUs, GPUs and NPUs.
NVIDIA research monorepo with four projects for long video generation, world models and robot world-action models, each in its own directory.