Mindket

Idol by Comfy-Org

Newversion 1 · updated

Overview

Turns an image into a 3D Gaussian splat inside ComfyUI, using repackaged files of a TripoSplat fine-tune. For people who build 3D results from pictures in ComfyUI workflows.

Idol by Comfy-Org is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page, a zip of its files. Nothing in it runs by itself.

At a glance

You need
  • ComfyUI installed on your computer
  • Python, for the Hugging Face download command
Runs in
  • ComfyUI, with the files placed in its models folders
  • Existing TripoSplat workflows
  • The Idol: Image to Gaussian Splat workflow template from the Comfy-Org workflow templates
  • Windows, macOS or Linux, wherever your ComfyUI runs
Start with
pip install -U huggingface_hub
hf download Comfy-Org/Idol

Platform

WindowsOperating system
macOSOperating system
LinuxOperating system

About

What it is

Idol is a set of repackaged model files for ComfyUI that turn an image into a 3D result. The model is a fine-tune of TripoSplat, so the TripoSplat workflows you already have can use it. The original model comes from the speridlabs/idol repository, and Comfy-Org packaged it for ComfyUI.

Who it suits

It suits ComfyUI users who want to make 3D scenes from single images.

IndustryWho uses itHow they use itExample
3D and games3D artistsMake a Gaussian splat from one pictureA concept image becomes a 3D scene you can look around
Product designDesignersPreview an object in 3D from a photoA product photo turned into a rough 3D view
ResearchML researchersRun an image-to-3D model inside a node workflowCompare Idol with TripoSplat in the same workflow

If you do not use ComfyUI, you can skip it.

Where to use it

  • ComfyUI, with the files placed in its models folders
  • Existing TripoSplat workflows
  • The Idol: Image to Gaussian Splat workflow template from the Comfy-Org workflow templates
  • Windows, macOS or Linux, wherever your ComfyUI runs
Use caseWhoWhat they get back
Image to Gaussian splat3D artistsA Gaussian splat made from one input image
Reusing TripoSplat workflowsComfyUI usersA workflow you already have, running with the Idol files

How to set it up

You need these first:

  • ComfyUI installed on your computer
  • Python, for the Hugging Face download command
  1. Download the model files with the hub tool.
  2. Copy each of the four files into its folder under the ComfyUI models folder, as the table below shows.
  3. Open the Idol workflow template, or an existing TripoSplat workflow, in ComfyUI.
pip install -U huggingface_hub
hf download Comfy-Org/Idol
Where each file goes inside the ComfyUI folder
FileFolder
idol_fp16.safetensorsmodels/diffusion_models
dino_v3_vit_h_bf16.safetensorsmodels/clip_vision
flux2-vae.safetensorsmodels/vae
idol_decoder_fp16.safetensorsmodels/vae

How to use it

Open the Idol: Image to Gaussian Splat workflow in ComfyUI, give it an image, and run it. The workflow uses the files you placed in the models folders.

Example

You load one photo of an object into the Idol workflow and run it. What comes back is a Gaussian splat of that object, the 3D result the workflow is named for.

For newcomers and experts

Idol is a pack of model files, not an app. A good first step is to place the four files in their folders, open the Idol workflow template, and try it with a single clear image.

Under the hood it is a fine-tune of TripoSplat, so it shares that model's workflows. The diffusion model and the decoder are stored in fp16, and the vision file is stored in bf16. A Flux 2 VAE file also sits in the vae folder.

  1. Open the Idol workflow in ComfyUI
  2. Give it one input image
  3. The Idol model files process the image
  4. A Gaussian splat comes out

Good to know

Idol is a repackaging for ComfyUI, so the model's own details live in the original repository.

  • Files: four safetensors files, split across the diffusion_models, clip_vision and vae folders
  • Compatibility: made to work with the existing TripoSplat workflows
  • License: Apache 2.0
  • Cost to run: none named
  • Data it reaches: the image you give it and the model files in your ComfyUI folders

README and files

README

Idol by Comfy-Org

Turns an image into a 3D Gaussian splat inside ComfyUI, using repackaged files of a TripoSplat fine-tune. For people who build 3D results from pictures in ComfyUI workflows.

Made by Comfy-Org on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face.

About

What it is: Idol is a set of repackaged model files for ComfyUI that turn an image into a 3D result. The model is a fine-tune of TripoSplat, so the TripoSplat workflows you already have can use it. The original model comes from the speridlabs/idol repository, and Comfy-Org packaged it for ComfyUI.

Who it suits: It suits ComfyUI users who want to make 3D scenes from single images.

Industry | Who uses it | How they use it | Example
3D and games | 3D artists | Make a Gaussian splat from one picture | A concept image becomes a 3D scene you can look around
Product design | Designers | Preview an object in 3D from a photo | A product photo turned into a rough 3D view
Research | ML researchers | Run an image-to-3D model inside a node workflow | Compare Idol with TripoSplat in the same workflow

If you do not use ComfyUI, you can skip it.

Where to use it:

  • ComfyUI, with the files placed in its models folders
  • Existing TripoSplat workflows
  • The Idol: Image to Gaussian Splat workflow template from the Comfy-Org workflow templates
  • Windows, macOS or Linux, wherever your ComfyUI runs
Use case | Who | What they get back
Image to Gaussian splat | 3D artists | A Gaussian splat made from one input image
Reusing TripoSplat workflows | ComfyUI users | A workflow you already have, running with the Idol files

How to set it up: You need these first:

  • ComfyUI installed on your computer
  • Python, for the Hugging Face download command
  1. Download the model files with the hub tool.
  2. Copy each of the four files into its folder under the ComfyUI models folder, as the table below shows.
  3. Open the Idol workflow template, or an existing TripoSplat workflow, in ComfyUI.
pip install -U huggingface_hub
hf download Comfy-Org/Idol
Title: Where each file goes inside the ComfyUI folder
File | Folder
idol_fp16.safetensors | models/diffusion_models
dino_v3_vit_h_bf16.safetensors | models/clip_vision
flux2-vae.safetensors | models/vae
idol_decoder_fp16.safetensors | models/vae

How to use it: Open the Idol: Image to Gaussian Splat workflow in ComfyUI, give it an image, and run it. The workflow uses the files you placed in the models folders.

Example: You load one photo of an object into the Idol workflow and run it. What comes back is a Gaussian splat of that object, the 3D result the workflow is named for.

For newcomers and experts: Idol is a pack of model files, not an app. A good first step is to place the four files in their folders, open the Idol workflow template, and try it with a single clear image.

Under the hood it is a fine-tune of TripoSplat, so it shares that model's workflows. The diffusion model and the decoder are stored in fp16, and the vision file is stored in bf16. A Flux 2 VAE file also sits in the vae folder.

Open the Idol workflow in ComfyUI
Give it one input image
The Idol model files process the image
A Gaussian splat comes out

Good to know: Idol is a repackaging for ComfyUI, so the model's own details live in the original repository.

  • Files: four safetensors files, split across the diffusion_models, clip_vision and vae folders
  • Compatibility: made to work with the existing TripoSplat workflows
  • License: Apache 2.0
  • Cost to run: none named
  • Data it reaches: the image you give it and the model files in your ComfyUI folders

On Hugging Face, 13 November 2026

  • Task: image to 3d
  • Library: diffusion-single-file
  • Downloads in the last 30 days: 52,939
  • Likes: 165
  • License: apache-2.0
  • Based on: speridlabs/idol
  • Tags: diffusion-single-file, comfyui, image-to-3d

Links

Get the model

Download its files with Hugging Face's command line:

pip install -U huggingface_hub
hf download Comfy-Org/Idol

Then follow the usage in its model card. Nothing in this kit runs by itself.

Files
README.md  4 KB

1 file, 4 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-12

Compare with similar

Listed Hugging Face models in the same category that share the most tags with this one.

ProductDownloads, last 30 daysLikesLicenseLast change
Idol by Comfy-Org This page52,939165apache-2.0
Idol Turbo by speridlabsShares image-to-3d · comfyui56557apache-2.0
Z-Image TurboShares diffusion-single-file · comfyui9,688,7281,030apache-2.0
MOSS-ImageShares diffusion-single-file · comfyui6,310,763666apache-2.0
Wan 2.1 for ComfyUIShares diffusion-single-file · comfyui4,888,8941,044apache-2.0
Qwen-Image-Edit by Comfy-OrgShares diffusion-single-file · comfyui1,898,657521apache-2.0
Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13.

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

Nanosaur2-670M

well9472 · Kit

Makes illustrations from text prompts with a small 670M-parameter diffusion model, run through ComfyUI. Built as a low-budget research model for people who wan…

New

Reads handwritten Arabic text from images and returns it as text. A fine-tuned PaddleOCR-VL model for people digitizing Arabic handwriting.

New

Makes video clips from a text prompt or an image plus text, at up to 720P and 24 frames per second, in the Diffusers format for a single consumer GPU.

New

Similar in Developer

See all in Developer

Wan2.2-S2V-14B

Wan-AI · Kit

Generates cinematic video of a character driven by an audio track, and can also lip-sync existing video. For video creators, animators and researchers with a l…

New

Wan2.2-Animate-14B

Wan-AI · Kit

Animates a character or swaps one into a video by copying a performer's movement and expressions, as one model from the Wan2.2 video family. Use only faces and…

New

d3-turbo

vllm-sr · Kit

Answers choice, yes-or-no, score, box and tracking questions about text, images or video in one call, with a probability for each answer. For people building r…

New