Downloads · 30 days
12.7K
38% of all-time downloads
TRELLIS.2-4B-GGUF
LocalAI-io/TRELLIS.2-4B-GGUF
TRELLIS.2-4B-GGUF is a image-to-3d model from LocalAI-io. 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 trellis. The card lists the license as mit.
GGUF (f16) conversions of microsoft/TRELLIS.2-4B for the trellis2cpp / ggml runtime — a CUDA-free, PyTorch-free C++ port of the TRELLIS.2 image-to-3D pipeline. Load these directly; no safetensors conversion or
All-time downloads
33.4K
Public
Repo size
15.7 GB
Likes
13
Public
Repo makeup
Click a slice to open those files.
.gguf15.7 GB · 100%
At a glance
- Task
- Image-to-3D
- Library
- trellis
- License
- mit
- Access
- Public
- Created
- Jul 16, 2026
- Updated
- Jul 16, 2026
- SHA
- 8d55d0a0
Base models
Model card
From the Hugging Face model README
TRELLIS.2-4B — GGUF (f16)
GGUF (f16) conversions of microsoft/TRELLIS.2-4B
for the trellis2cpp /
ggml runtime — a CUDA-free, PyTorch-free C++ port of
the TRELLIS.2 image-to-3D pipeline. Load these directly; no safetensors conversion or
Python is needed at inference.
Files
| File | Pipeline stage | Source safetensors |
|---|---|---|
ss_flow_f16.gguf | Sparse-structure flow (64³ occupancy) | ss_flow_img_dit_1_3B_64_bf16 |
slat_flow_f16.gguf | Shape-SLAT flow, 512 fine | slat_flow_img2shape_dit_1_3B_512_bf16 |
slat_flow_1024_f16.gguf | Shape-SLAT flow, 1024 cascade | slat_flow_img2shape_dit_1_3B_1024_bf16 |
shape_dec_f16.gguf | Shape decoder → dual-grid fields | shape_dec_next_dc_f16c32_fp16 |
shape_enc_f16.gguf | Shape encoder (re-encode for texture flow) | shape_enc_next_dc_f16c32_fp16 |
tex_dec_f16.gguf | PBR texture decoder | tex_dec_next_dc_f16c32_fp16 |
tex_slat_flow_512_f16.gguf | Texture-SLAT flow, 512 | slat_flow_imgshape2tex_dit_1_3B_512_bf16 |
tex_slat_flow_1024_f16.gguf | Texture-SLAT flow, 1024 | slat_flow_imgshape2tex_dit_1_3B_1024_bf16 |
The image conditioning encoder (DINOv3) and the sparse-structure decoder are published separately: dinov3-vitl16-pretrain-lvd1689m-GGUF and TRELLIS-image-large-GGUF.
Conversion
Converted from the upstream bf16/fp16 safetensors to GGUF f16 (ggml ftype 1) with
the convert_*_to_gguf.py scripts in the trellis2cpp project. No weights were retrained
or modified — this is a format conversion only.
License
MIT, inherited from microsoft/TRELLIS.2-4B.
See LICENSE. Copyright (c) Microsoft Corporation.
- Task
- Image-to-3D
- Library
- trellis
- License
- mit
- Created
- Jul 16, 2026
- Updated
- Jul 16, 2026