Skip to content
ComfyUI Wiki · 工作流与开源创作分享·· 2026-10-16PicksAI score65

OpenMOSS releases 5B text-to-image model MOSS-Image, with same-day ComfyUI support

Text to image

AI briefing

OpenMOSS released the 5B text-to-image model family MOSS-Image on October 16, including two checkpoints: the pretrained version MOSS-Image-Base and the general-user-oriented MOSS-Image-SFT.

Why it matters

The 5B text-to-image model focuses on Chinese and English text rendering and JSON layout control, with same-day support from ComfyUI core nodes, which can be used to evaluate layout-oriented image generation workflows.

Open original comfyui-wiki.com