TRELLIS.2-4B-diffusers-3d

suvadityamuk/TRELLIS.2-4B-diffusers-3d

TRELLIS.2-4B-diffusers-3d is a image-to-3d model from suvadityamuk. Use it for the image-to-3d task on the model card, and read the license before you ship it in a product. It is set up for diffusers. The card lists the license as mit.

microsoft/TRELLIS.2-4B converted into a diffusers-3d pipeline.

Downloads · 30 days

100% of all-time downloads

All-time downloads

Public

Repo size

15 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors15 GB · 100%

At a glance

Task: Image-to-3D
Library: diffusers
License: mit
Access: Public
Created: Sep 16, 2026
Updated: Sep 16, 2026
SHA: 20af78a7

Base models

microsoft/TRELLIS.2-4B

Model card

From the Hugging Face model README

TRELLIS.2-4B for diffusers-3d

microsoft/TRELLIS.2-4B converted into a diffusers-3d pipeline.

The DINOv3 image conditioner is not included. Its weights are gated under Meta's DINOv3 License, so this repository ships every TRELLIS.2 component except conditioner/; you accept the license on facebook/dinov3-vitl16-pretrain-lvd1689m and pass the conditioner in when loading.

Install

pip install git+https://github.com/suvadityamuk/diffusers.git
pip install "git+https://github.com/suvadityamuk/diffusers.git#subdirectory=packages/diffusers-3d"

diffusers-3d runs every network in plain PyTorch on CPU or GPU. Rendering Gaussian splats needs the optional gsplat backend; meshing and PBR export for TRELLIS.2 need the compiled O-Voxel runtime (see the package docs).

Usage

import torch
from diffusers_3d import AutoPipelineForImageTo3D, ImageCondition, Trellis2Dinov3Conditioner

conditioner = Trellis2Dinov3Conditioner.from_dinov3_pretrained("facebook/dinov3-vitl16-pretrain-lvd1689m")
pipeline = AutoPipelineForImageTo3D.from_pretrained(
    "suvadityamuk/TRELLIS.2-4B-diffusers-3d", conditioner=conditioner, dtype=torch.bfloat16
).to("cuda")
output = pipeline(ImageCondition(image=rgba), pipeline_type="512")
ovoxel = output.objects[0]  # OVoxelAsset: dual-grid surface with PBR channels

pipeline_type selects the released preset (512, 1024, 1024_cascade, 1536_cascade). Meshing the O-Voxel (OVoxelBackend.to_mesh) and baking the PBR GLB (pipeline.postprocess_ovoxel(..., output_format="glb")) need the compiled O-Voxel runtime and its research-licensed nvdiffrast dependency; see the package docs.

Components

Folder	Class	Released file
`conditioner`	`Trellis2Dinov3Conditioner`	not included (gated `facebook/dinov3-vitl16-pretrain-lvd1689m`)
`sparse_structure_flow_model`	`Trellis2SparseStructureFlowModel`	`ss_flow_img_dit_1_3B_64_bf16`
`sparse_structure_decoder`	`Trellis2SparseStructureDecoder`	`ss_dec_conv3d_16l8_fp16` (from TRELLIS-image-large)
`shape_slat_flow_model`	`Trellis2SLatFlowModel`	`slat_flow_img2shape_dit_1_3B_512_bf16`
`shape_slat_flow_model_1024`	`Trellis2SLatFlowModel`	`slat_flow_img2shape_dit_1_3B_1024_bf16`
`shape_slat_decoder`	`Trellis2ShapeDualGridDecoder`	`shape_dec_next_dc_f16c32_fp16`
`texture_slat_flow_model`	`Trellis2SLatFlowModel`	`slat_flow_imgshape2tex_dit_1_3B_512_bf16`
`texture_slat_flow_model_1024`	`Trellis2SLatFlowModel`	`slat_flow_imgshape2tex_dit_1_3B_1024_bf16`
`pbr_decoder`	`Trellis2PBRSparseDecoder`	`tex_dec_next_dc_f16c32_fp16`

Provenance

Converted with diffusers-3d-convert-trellis2 from diffusers-3d 0.1.0.dev0 against TRELLIS.2 revision 75fbf0183001ed9876c8dbb35de6b68552ee08bd. Weight values are unchanged.

License and attribution

TRELLIS.2 weights and architecture: MIT License, Copyright (c) Microsoft Corporation. The sparse-structure decoder comes from TRELLIS (MIT). No DINOv3 weights are redistributed. Not affiliated with or endorsed by Microsoft or Meta.

Task: Image-to-3D
Library: diffusers
License: mit
Created: Sep 16, 2026
Updated: Sep 16, 2026