Amid the controversy surrounding FLUX3, a new foundational model was quietly released. Its standout features include support for regional prompts and improved text rendering capabilities, alongside extremely fast runtime. This is Moss-image. ComfyUI added support for it within 24 hours of its launch, though the integration has not yet been merged into the main branch. Users who want to try it need to update ComfyUI to the latest version available on GitHub.

The model address is here

Comfy-Org/MOSS-Image · Hugging Face

ComfyUI has not provided a workflow at present, so I built one manually. You can get it from the attachment.

Notes:

The model has two versions: base and sft. The sft version is enhanced for text rendering. The int8 quantized version is preferred.

Since the model has not been distilled, the CFG scale can be increased appropriately to 2~3, and negative prompts are effective.

The built-in regional prompt node of ComfyUI runs under the Nodes2.0 environment, and you need to enable Nodes2.0 to use this feature.

OK! That's all! I hope you like it!

If you like my sharing and want to get more content and resources related to AI image generation, welcome to join my private community.👇👇👇

NO8D-AI-Workshop

🫡May the AI-power be with you, see you soon!🫡

Disclaimer: My model or workflow is solely for learning and communication purposes and shall not be used for any commercial activities. All responsibilities for the generated content, including its legality, ethical implications, or potential misuse, are borne entirely by the user. By using this model or workflow, you agree to this principle. Please use it responsibly within the scope of local laws.