MoGe

MoGe

Estimates 3D geometry and depth from a single image, as repackaged MoGe model files for ComfyUI with workflows for meshes and depth maps. For ComfyUI users who work with 3D and depth.

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: MoGe is a monocular geometry estimation model, which means it works out 3D shape and depth from one image. This repository holds four MoGe model files repackaged by Comfy-Org for ComfyUI. The card links four workflow templates that turn images into meshes, panoramas into meshes, or depth estimates.

Who it suits: ComfyUI users who want depth or 3D geometry from still images or panoramas, such as artists, 3D hobbyists and people preparing inputs for other pipelines.

Title: Who uses it, and how
Industry | Who uses it | How they use it | Example
3D art | Artists | Turn a picture into a rough mesh | Run the perspective workflow on a photo of a room
Virtual scenes | Scene builders | Turn a panorama into a mesh | Run the panorama workflow on a wide panoramic photo
Image editing | Compositors | Make a depth map to guide edits | Use the depth workflow on a portrait
Research | Vision researchers | Compare the MoGe versions on the same images | Run MoGe 2 and MoGe 3 on one photo

Skip it if you do not use ComfyUI, because these files are packaged for that program.

Where to use it:

• ComfyUI, on Windows, macOS or Linux
• The four workflow templates the card lists
• Windows, macOS or Linux for downloading the files, with Python

Use case | Who | What they get back
Perspective to mesh | 3D artists | A mesh built from a normal photo
Panorama to mesh | Scene builders | A mesh built from a panoramic image
Depth estimation | Compositors | A depth result shown before and after in the template

How to set it up: You need ComfyUI installed, plus Python and pip to download. Then you copy the model files into the geometry_estimation folder inside ComfyUI's models directory.

• ComfyUI
• Python and pip
• Disk space for the model files

1. Install the Hugging Face hub library.
2. Download the repository.
3. Copy the files into the geometry_estimation folder, then open a workflow.

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

Title: The four model files
File | Version
moge_1_vitl_fp16 | MoGe 1, large model
moge_2_vitl_normal_fp16 | MoGe 2, large model with normals
moge_3_vitg_fp16 | MoGe 3, giant model
moge_3_vitl_fp16 | MoGe 3, large model

How to use it: Open one of the card's workflow templates in ComfyUI, load an image or panorama, and run it. Each template points at one of the model files, so check that the file you want is in place.

Tool | What it does
Perspective Geometry Estimation | Turns a normal image into geometry and a mesh
Panorama to Mesh | Builds a mesh from a panoramic image
Depth Estimation | Produces a depth result, with a before and after view
Geometry Estimation with MoGe 3 | Runs the newer MoGe 3 models

Example: You load a photo of a room into the perspective workflow and run it. The workflow returns the room's estimated geometry as a mesh. The card shows before and after pictures for the depth template.

For newcomers and experts: If you are new, learn the basics of ComfyUI first, then run the depth workflow on one photo, since it is the simplest output to check by eye.

For people in the field, the repository packages MoGe 1, MoGe 2 with normals and MoGe 3 in two sizes, all in fp16. Their original sources are Microsoft's MoGe project and the Ruicheng model repositories. The card gives no benchmark figures.

Download the repository
Copy files into geometry_estimation
Open a MoGe workflow in ComfyUI
Load an image and run it
Get a mesh or depth result

Good to know:

• License: MIT
• Cost to run: none named, you use your own computer
• Data it reaches: it downloads files from Hugging Face, and the card names nothing else it touches
• Not for: use outside ComfyUI, since the card only describes that setup
• The card gives no sizes in gigabytes and no hardware notes

On Hugging Face, 13 November 2026

• Library: diffusion-single-file
• Downloads in the last 30 days: 176,155
• Likes: 66
• License: mit
• Based on: Ruicheng/moge-vitl
• Tags: diffusion-single-file, comfyui

Links

• Model: https://huggingface.co/Comfy-Org/MoGe
• Files: https://huggingface.co/Comfy-Org/MoGe/tree/main

Get the model

Download its files with Hugging Face's command line:

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

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