Downloads · 30 days
98
75% of all-time downloads
Pixal3D-Comfy
chibifire/Pixal3D-Comfy
Pixal3D-Comfy is a machine learning model from chibifire. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for diffusion-single-file. The card lists the license as mit.
Pixal3D model files repackaged for ComfyUI, in the ComfyUI models folder layout.
All-time downloads
130
Public
Repo size
36.4 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors36.4 GB · 100%
At a glance
- Library
- diffusion-single-file
- License
- mit
- Access
- Public
- Created
- Sep 6, 2026
- Updated
- Oct 11, 2026
- SHA
- 2237a20a
Base models
Model card
From the Hugging Face model README
Pixal3D-Comfy
Pixal3D model files repackaged for ComfyUI, in the ComfyUI models folder layout.
What it is for
This repo holds the Pixal3D weights repackaged as single safetensors files for ComfyUI.
It has standard and multiview diffusion models in bf16 and int8 form, a DINO vision encoder, and the TRELLIS.2 shape and texture VAEs.
The folders match the ComfyUI models subfolders, so a download drops into place.
It also holds a JSON test file.
Provenance
- Upstream model: TencentARC/Pixal3D, the
base_modelin the header.
Use
hf download chibifire/Pixal3D-Comfy --local-dir ComfyUI/models --include "diffusion_models/*" "clip_vision/*" "vae/*"
This puts the diffusion models in models/diffusion_models, the vision encoder in models/clip_vision and the VAEs in models/vae.
For model details, read the upstream model card.
Licence
MIT, the upstream licence. Terms: https://opensource.org/license/mit.
- Library
- diffusion-single-file
- License
- mit
- Created
- Sep 6, 2026
- Updated
- Oct 11, 2026