MOSS-Image-Base

What is MOSS-Image-Base

MOSS-Image series text-to-image foundation models released by OpenMOSS. Supports natural language prompts, tag sequences and JSON layouts to control text placement and object arrangement. Apache-2.0 open source.

Key features

• Supports natural language prompts, tag sequences and JSON layouts to control text placement and object arrangement
• Supports native ComfyUI plugin integration, installable via ComfyUI Manager
• Loads transformer and text encoder in bfloat16 precision, VAE in float32

Who it is for

Researchers, developers and ComfyUI users who need to generate controllable layout images. Suitable for text-to-image tasks involving text and structured layouts.

How to get started

1. Visit https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base to download the model
2. Run DiffusionPipeline.from_pretrained("OpenMOSS-Team/MOSS-Image-Base", custom_pipeline="pipeline_moss_image", trust_remote_code=True)
3. Search for MOSS-Image in the ComfyUI Manager to install the native plugin

Compiled from public descriptions on the tool's official website / README / launch page. Pricing and user numbers have not been verified; please check the official website. Some sections were drafted with AI assistance. Written in Chinese and English; other languages are machine-translated.