Learn & explore
MOSS-Image Guides & Research
From your first poster to a local setup: learn prompting and layout, choose a model, and inspect real experiment results.
Practical guides
- How do I make an image with MOSS-Image for free?
Compare ways to use MOSS-Image, understand free-image limits, and follow four short steps to create, check and download an image online.
- How do I use MOSS-Image in ComfyUI?
Set up MOSS-Image in ComfyUI step by step: which build you need, the three files to download, where they go and which workflow to load.
- How much VRAM does MOSS-Image need?
How much VRAM MOSS-Image needs, bf16, int8 and w4a8 file sizes, GGUF options, memory modes, and what Mac and low-VRAM users can do.
- Should I use MOSS-Image Base, SFT or Turbo?
The difference between MOSS-Image Base and SFT, what the community 4-step and 8-step Turbo versions trade off, and which one to pick.
- How do I write prompts for MOSS-Image?
How to write MOSS-Image prompts that render text correctly: quoting words, line breaks, negative prompts, JSON layouts and when a more detailed prompt helps.
Experiments & evidence
- MOSS-Image text and design test: how to get the words right, and how to edit safely
Can MOSS-Image get the words right on menus, posters and labels? We compare short prompts with automatic expansion across 8 scenes and test text edits, added text and image-editing models.