How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("OzzyGT/hunyuan3d_2mini_turbo", dtype=torch.bfloat16, device_map="cuda")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

Hunyuan3D-2mini Turbo

Front, right, back and left views of a shape generated with the turbo checkpoint

Generated with the code below (seed 42, 30 steps, octree resolution 380).

This is tencent/Hunyuan3D-2mini, the step-distilled turbo checkpoint (hunyuan3d-dit-v2-mini-turbo), repacked into the Diffusers format with the original fp16 weights. It runs as a Modular Diffusers pipeline with the Hunyuan3D-2mini custom blocks and generates an untextured 3D shape from a single image.

Note: You need diffusers>=0.41.0 and scikit-image. The example also uses trimesh to save the mesh.

Usage

The custom blocks load the model code shipped in this repository, hence trust_remote_code and trust_components_code.

Note: both flags run Python code downloaded from the Hub: trust_remote_code=True runs the code of the blocks repository, and trust_components_code=True runs the code shipped in the components_repo. Only use them with repositories you trust, and review their code before passing different ones.

import torch
import trimesh
from diffusers import ModularPipelineBlocks
from diffusers.utils import load_image


blocks = ModularPipelineBlocks.from_pretrained(
    "OzzyGT/hunyuan3d_2mini_custom_blocks",
    trust_remote_code=True,
    components_repo="OzzyGT/hunyuan3d_2mini_turbo",
    trust_components_code=True,
)
pipe = blocks.init_pipeline()
pipe.load_components(dtype=torch.float16)
pipe.to("cuda")

image = load_image(
    "https://huggingface.co/datasets/OzzyGT/diffusers-examples/resolve/main/hunyuan3d_2mini/20261016231355.png",
    convert_method=lambda image: image.convert("RGBA"),
)
mesh = pipe(
    image=image,
    num_inference_steps=30,
    octree_resolution=380,
    generator=torch.Generator().manual_seed(42),
    output="meshes",
)[0]
trimesh.Trimesh(mesh["vertices"], mesh["faces"]).export("shape.glb")

The input is an image of a single object with a transparent background.

License

This model is released under the Tencent Hunyuan 3D 2.0 Community License, the same as the original; see also the NOTICE. The license does not apply in the European Union, the United Kingdom and South Korea. Powered by Tencent Hunyuan.

Downloads last month
1
Safetensors
Model size
0.6B params
Tensor type
F16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for OzzyGT/hunyuan3d_2mini_turbo

Finetuned
(5)
this model

Space using OzzyGT/hunyuan3d_2mini_turbo 1