Report
freeTrialImage.bannerPity
freeTrialImage.upgradeUnlock
- ✓freeTrialImage.benefitHd
- ✓freeTrialImage.benefitWatermark
- ✓freeTrialImage.benefitUnlimited
Comfyui Moss Image
Comfyui Moss Image brings box-guided JSON prompts to your ComfyUI canvas, so every element lands exactly where you planned it. Try it free.
AI Tools
Explore AI tools
Free AI Image
Truly Free AI Image Generator
Image Upscaler
Enhance image resolution effortlessly
AI Color Changer
Transform photo colors in seconds
AI Room Designer
Design your dream room with AI-powered 3D rendering
Skin Enhancer
Enhance skin texture naturally with AI
Face Editor
Enhance face texture naturally with AI
AI Clothes Changer
Meet Comfyui Moss Image: A Compact Layout Engine for ComfyUI
A 5B text-to-image family that reads structured JSON, places objects by box coordinates, and keeps lettering legible in anime or photo looks.
- Structured JSON, Not GuessworkEvery bounding box inside the JSON prompt pins an object's position and size — rendered text blocks and scene elements included.
- Long Prompts on a Small BackboneInstruction up to 4K tokens runs on a compact 5B core, and the Base and SFT checkpoints share one sampling interface.
- Legible Text in Anime and Photo StylesLettering stays crisp across illustration, photography, and design work, with anime and photorealistic looks both well covered.
Set Up Comfyui Moss Image in Three Steps
From plugin install to a finished boxed layout render, in three short steps.
Comfyui Moss Image — Quick Answers
Answers to the questions people ask before running boxed layouts in ComfyUI.
Is the official ComfyUI repository required?
Not at all — you find the native plugin by searching MOSS-Image in ComfyUI Manager, and the work of merging it into the official repository is still ongoing.
How is SFT different from MOSS-Image-Base?
The two checkpoints share an architecture and sampling interface; in the reference example, SFT produced tighter text grouping and a cleaner whiteboard composition.
What role do bounding boxes play in a prompt?
A box defines where an element sits and how large it is; editing the text, the element descriptions, or the boxes themselves reshapes both content and composition.
How long can prompts be, and which styles work?
Instruction can run up to 4K tokens, and the model covers anime and photorealistic looks across text, illustration, photography, and visual design.
Which settings help when VRAM is limited?
Advanced settings list GPU resident, Component, and Sequential modes. Pick Sequential or scale the image down when memory is tight; note that Compile DiT is off by default and cannot be used in Sequential.
Is fine-tuning allowed under Apache-2.0?
Yes — the 5B backbone takes fine-tuning for custom tasks and domains, and the weights ship under Apache-2.0, while pipeline code and third-party components keep their own licenses.
Run Your First Boxed Layout With Comfyui Moss Image
Grab the plugin in ComfyUI Manager, load the sample workflow, and push a boxed JSON prompt through either Base or SFT weights.