MOSS-Image: Text-to-Image Meets Layout-to-Image
freeTrialImage.bannerPity

Report

Output
Preview of your layout

freeTrialImage.bannerPity

freeTrialImage.upgradeUnlock

  • ✓freeTrialImage.benefitHd
  • ✓freeTrialImage.benefitWatermark
  • ✓freeTrialImage.benefitUnlimited

Comfyui Moss Image

Explore Comfyui Moss Image in ComfyUI: build JSON layouts with bounding boxes, switch between Base and SFT weights, and get sharp on-image text.

AI Tools

Explore AI tools

placeholder hero

Comfyui Moss Image: Precise Layout Control for ComfyUI

A compact 5B text-to-image model built for structured JSON prompts, reliable text drawing, and both anime and photoreal results.

  • Bounding-Box JSON Prompts
    Structured JSON entries and bounding boxes define where objects, text, and scene details appear, giving you direct control over composition.
  • Long Prompts on a Small Backbone
    Feed prompts up to 4K tokens into a 5B model, with Base and SFT checkpoints sharing the same sampling workflow.
  • Readable Text, Two Visual Styles
    Get legible on-image lettering plus anime and photoreal looks that suit illustrations, photos, and design layouts.

Set Up Comfyui Moss Image in Three Steps

Install the plugin, load a JSON layout prompt, then render while keeping VRAM in check.

FAQ

Comfyui Moss Image: Common Questions

Answers about plugin setup, checkpoint choices, bounding boxes, prompt limits, VRAM modes, and licensing.

1

Do I Need the Official ComfyUI Repository?

Not at all. You can find the native plugin by searching MOSS-Image in ComfyUI Manager; work to add it to the official repository is still underway.

2

How Do Base and SFT Checkpoints Differ?

They use the same architecture and sampling interface. In the reference example, SFT produced tighter text grouping and a cleaner whiteboard composition.

3

What Role Do Bounding Boxes Play?

Every box sets an element's position and size. You can adjust the text, element notes, and boxes to reshape both the message and the composition.

4

What Prompt Length and Styles Can I Use?

You can supply up to 4K tokens and work in anime or photorealistic styles for text, illustrations, photography, and visual design.

5

Which Memory Mode Should Low-VRAM Users Pick?

Advanced settings include GPU resident, Component, and Sequential modes. Choose Sequential or reduce image size when memory is low; Compile DiT stays off by default and is unavailable in Sequential.

6

Can I Fine-Tune the Apache-2.0 Weights?

Yes. The 5B backbone can be fine-tuned for custom tasks and domains, and the model weights use Apache-2.0; pipeline code and third-party components keep their own licenses.

Start Using Comfyui Moss Image Today

Bring the plugin into ComfyUI Manager, launch the starter workflow, and compare Base and SFT on your first JSON-driven layout.