Hunyuan3D-2
Generates high-resolution textured 3D assets from images or text with separate shape and texture models.
About
Tencent's Hunyuan3D-2 splits textured 3D asset generation into two stages: Hunyuan3D-DiT, a flow-based diffusion transformer that generates the mesh from an image or text prompt, and Hunyuan3D-Paint, which synthesizes high-resolution textures for it. The family covers a 3.0B parameter shape model and 1.3B texture model in version 2.1, a 0.6B mini variant, a 1.1B multiview model, and turbo builds for fast inference, with the June 2025 release of 2.1 also open-sourcing training code and a PBR texture pipeline. Shape generation runs in about 6 GB of VRAM while the full shape plus texture pipeline wants 16 GB, and the repo ships a Gradio app, an HTTP API server, and a Blender addon alongside pip-based installation. At around 14.4k GitHub stars it is the most adopted open 3D asset generator. Weights are distributed under the Tencent Hunyuan 3D 2.0 Community License, which permits commercial use below 1 million monthly active users but excludes the EU, UK, and South Korea, and an official hosted version runs at 3d.hunyuan.tencent.com.
Reviews (0)
Leave a Review
No reviews yet. Be the first to review!
Details
- Category
- 3D Model Generation
- Price
- Free
- Platform
- Local/Desktop
- Difficulty
- Intermediate (3/5)
- License
- Tencent Hunyuan 3D 2.0 Community License
- Minimum VRAM
- 6 GB
- Added
- Jul 29, 2026
Related Tools
CUDA-accelerated Gaussian splatting rasterization library with Python bindings from the Nerfstudio team.
Generates explorable, mesh-exportable 3D worlds from text or image prompts via panoramic proxies.
Roblox's research foundation model that generates 3D shapes from text prompts for game assets.
Reconstructs 3D shape, texture, and layout of objects from a single cluttered real-world photo.
Turns a single image into a textured, UV-unwrapped 3D mesh in under a second.
Reconstructs dense 3D scenes from uncalibrated image collections without known camera poses.