Hunyuan3D 2.1
Tencent's first fully open-source AI 3D modeling tool
Tencent's first fully open-source AI 3D modeling tool with full-pipeline 3D generation and PBR materials.
Image generation and computer vision — detection, editing, 3D.
Key features
- Production-ready 3D asset generation: image-to-shape (3.3B) plus PBR texture synthesis
- Fully open-source framework with full model weights and training code
- PBR pipeline with physics-grounded material simulation (metallic reflections, subsurface
- Outperforms open and closed-source baselines on textured 3D asset quality benchmarks
- Diffusers-like API; 10GB VRAM for shape, 21GB for texture, 29GB combined
Good for
- Generating production-ready textured 3D assets from images
- Game and XR developers needing 3D asset pipelines (ComfyUI, Unity
Stack & tags
PythonPyTorch3d-generationtextured-3dpbropen-weightsimage-to-3dgame-assetstencent
Quick start
pip install torch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1 --index-url https://download.pytorch.org/whl/cu124
pip install -r requirements.txtFrom the project README on GitHub.
Similar open-source projects
FLUX.2 klein
★ 25,996Black Forest Labs' sub-second image generation — generate and edit in <0.5s
DiffSynth-Studio
★ 13,200The framework behind Qwen-Image-i2L — the first "Image→LoRA" visual generation system