Skip to content
ModelListedExternal
Browse files

Dvd

This repository hosts the pretrained DVD checkpoints for Discrete Voxel Diffusion for 3D Generation and Editing.

KindOtherVersionImg BSP ftBgrade

Model source

Kind
Other
File size
1.50–1.55 GB
Version
vImg BSP ft
Updated
Oct 10, 2026
Source
Hugging Face

Source description

DVD: Discrete Voxel Diffusion for 3D Generation and Editing

This repository hosts the pretrained DVD checkpoints for Discrete Voxel Diffusion for 3D Generation and Editing.

DVD is a discrete diffusion pipeline that generates or edits a pure 64^3 voxel grid. The voxel output can be used directly, exported as a cubified voxel mesh, or passed as the sparse-structure anchor for TRELLIS stage 2 to produce final mesh and 3D Gaussian assets.

Checkpoints

VariantFilesUse case
Image generationdvd_img.json, dvd_img.safetensorsimage-conditioned voxel generation
Image editingdvd_img_BSP_ft.json, dvd_img_BSP_ft.safetensorsimage-conditioned voxel editing
Text generationdvd_text.json, dvd_text.safetensorstext-conditioned voxel generation
Text editingdvd_text_BSP_ft.json, dvd_text_BSP_ft.safetensorstext-conditioned voxel editing

Minimal Usage

from dvd import DVDImageToVoxelPipeline, DVDTextToVoxelPipeline

repo_id = "Zhengrui/dvd"

image_dvd = DVDImageToVoxelPipeline.from_pretrained(repo_id, variant="base", device="cuda")
image_edit_dvd = DVDImageToVoxelPipeline.from_pretrained(repo_id, variant="bsp", device="cuda")

text_dvd = DVDTextToVoxelPipeline.from_pretrained(repo_id, variant="base", device="cuda")
text_edit_dvd = DVDTextToVoxelPipeline.from_pretrained(repo_id, variant="bsp", device="cuda")

Example image-conditioned generation:

from PIL import Image
from dvd import DVDImageToVoxelPipeline, export_cubified_voxels

image = Image.open("assets/example_image/T.png")
dvd = DVDImageToVoxelPipeline.from_pretrained("Zhengrui/dvd", variant="base", device="cuda")
voxels = dvd.sample_voxels(image, seed=42, steps=256)
export_cubified_voxels(voxels, "dvd_voxels.glb")

Voxel Convention

DVD stores voxel outputs as:

samples: [B, D, H, W]
coords:  [N, 4] as [batch, x, y, z]

Saved .npy coordinate files usually omit the batch dimension and can be loaded back with as_voxel_output(coords, resolution=64).

Notes

These files are the DVD voxel-stage checkpoints. TRELLIS stage-2 checkpoints are loaded separately from the official TRELLIS model repositories, such as microsoft/TRELLIS-image-large and microsoft/TRELLIS-text-large.

Citation

@misc{xiang2026dvddiscretevoxeldiffusion,
      title={DVD: Discrete Voxel Diffusion for 3D Generation and Editing},
      author={Zhengrui Xiang and Jiaqi Wu and Fupeng Sun and Heliang Zheng and Yingzhen Li},
      year={2026},
      eprint={2605.07971},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2605.07971},
}

Sources

1 source
Hugging FacePrimaryVersion bfc4d6873d1283e342c1a68b61a4b3bb31de1d6b
huggingface.co/Zhengrui/dvd/tree/bfc4d6873d1283e342c1a68b61a4b3bb31de1d6b

Verified Oct 10

Download files

Model artifacts

4 artifacts

dvd_img_BSP_ft.safetensors

safetensors · 1.55 GB · SHA-256 6e053b0fdab3…a307 · Hugging Face

Download file

dvd_img.safetensors

safetensors · 1.55 GB · SHA-256 b73ba077d29a…7b69 · Hugging Face

Download file

dvd_text_BSP_ft.safetensors

safetensors · 1.50 GB · SHA-256 32bf1fb84c39…74a4 · Hugging Face

Download file

dvd_text.safetensors

safetensors · 1.50 GB · SHA-256 1b928adb953b…46db · Hugging Face

Download file

Source excerpts

2 excerpts

--- license: mit library_name: pytorch tags: - 3d-generation - voxel-generation - 3d-editing - diffusion - image-to-3d - text-to-3d - trellis pipeline_tag: image-to-3d --- # DVD: Discrete Voxel Diffusion for 3D Generation and Editing This repository hosts the pretrained DVD checkpoints for **Discrete Voxel Diffusion for 3D Generation and Editing**. DVD is a discrete diffusion pipeline that generates or edits a pure `64^3` voxel grid. The voxel output can be used directly, exported as a cubified voxel mesh, or passed as the sparse-structure anchor for [TRELLIS](https://github.com/microsoft/TRELLIS) stage 2 to produce final mesh and 3D Gaussian assets. - Paper: [arXiv:2605.07971](https://arxiv.org/abs/2605.07971) - Spaces: [DVD Image](https://huggingface.co/spaces/Zhengrui/dvd-image), [DVD Text](https://huggingface.co/spaces/Zhengrui/dvd-text) ## Checkpoints | Variant | Files | Use case | | --- | --- | --- | | Image generation | `dvd_img.json`, `dvd_img.safetensors` | image-conditioned voxel generation | | Image editing | `dvd_img_BSP_ft.json`, `dvd_img_BSP_ft.safetensors` | image-conditioned voxel editing | | Text generation | `dvd_text.json`, `dvd_text.safetensors` | text-conditioned voxel generation | | Text editing | `dvd_text_BSP_ft.json`, `dvd_text_BSP_ft.safetensors` | text-conditioned voxel editing | ## Minimal Usage ```python from dvd import DVDImageToVoxelPipeline, DVDTextToVoxelPipeline repo_id = "Zhengrui/dvd" image_dvd = DVDImageToVoxelPipeline.from_pretrained(repo_id, variant="base", device="cuda") image_edit_dvd = DVDImageToVoxelPipeline.from_pretrained(repo_id, variant="bsp", device="cuda") text_dvd = DVDTextToVoxelPipeline.from_pretrained(repo_id, variant="base", device="cuda") text_edit_dvd = DVDTextToVoxelPipeline.from_pretrained(repo_id, variant="bsp", device="cuda") ``` Example image-conditioned generation: ```python from PIL import Image from dvd import DVDImageToVoxelPipeline, export_cubified_voxels image = Image.open("assets/example_image/T.png") dvd = DVDImageToVoxelPipeline.from_pretrained("Zhengrui/dvd", variant="base", device="cuda") voxels = dvd.sample_voxels(image, seed=42, steps=256) export_cubified_voxels(voxels, "dvd_voxels.glb") ``` ## Voxel Convention DVD stores voxel outputs as: ```text samples: [B, D, H, W] coords: [N, 4] as [batch, x, y, z] ``` Saved `.npy` coordinate files usually omit the batc...

Source context: 0 downloads · 1 likes · Pipeline image-to-3d · Library pytorch · Repo Zhengrui/dvd