This workflow is based on the information found via ComfyUI's MOSS text-to-image templates and the official MOSS-Image user guide. The ComfyUI template is fine if you want something really simple to use. My workflow is not recommended for brand new ComfyUI users.
This workflow has the following functions/features:
Text to Image
Multiple Lora Loading
Wildcards
Prompt Reformatting
Optional Upscaling (if needed)
Save Image with full resource metadata in CivitAI readable format.
Please be aware that this workflow was designed to be used with the classic ComfyUI canvas and is not Nodes 2.0 friendly by default.
There is a list of the custom nodes used on the workflow canvas with links since ComfyUI Manager is not always reliable.
Support Policy
Refer to the official MOSS-Image user guide or the CivitAI MOSS-Image model page for guidance on how to use the model.
These workflows are provided as-is and without guaranteed support.
I do not provide support for third-party applications, modified versions, redistributed copies, reposts, bundles, or installations obtained from unofficial sources.
Inclusion of these workflows in another project does not imply my endorsement, involvement, or responsibility for that project.
Disclaimer
Responsibility for generated content falls on the user of these workflows. I do not control how people prompt or what they use.
Downloading/installation of ComfyUI and/or any related custom nodes for these workflows are at the installer's own risk.
Description
Replaced the v1 JSON with a fixed version for the missing image input on the SeedVR2 switch at 730PM EDT on 25 October 2026.
This is my entry workflow for MOSS Image.
The funky model patch nodes are there since both the ComfyUI template and the official MOSS Image repo mention using them.
There are some basic info notes on the workflow about them, but when in doubt refer to the official repo user guide or the CivitAI model page for MOSS Image OR ask your preferred AI (ChatGPT, Gemni, Claude, etc.)
The settings saved to the workflow are what I use.
I recommend experimenting to find your preferred results since I am not the guru of ComfyUI or MOSS Image.
With a Turbo lora, I am able to generate a 2x SDXL sized image in under 30 seconds on a 5060ti 16GB.