A quick MOSS-Image news round-up - 22nd October 2026
retained text
Another quick MOSS-Image news and goodies round-up, for those who may have missed some items. -> ComfyUI-Spectrum-MOSS's readme now has important new guidance on settings for those using the *euler* or *er-sde* samplers, advice which was added a few hours ago. All *er-sde* users should also be sure to update to the latest v0.8.1 version, as this has... "Updated the ER-SDE bridge to the new, improved logic" and the readme adds the new settings. I saw a definite speed-up of perhaps 20% on a 3060 12Gb card, when using the new update, *er-sde* and the new settings, compared to the earlier Spectrum-MOSS version. https://github.com/Rkkss/ComfyUI-Spectrum-MOSS -> A minor MOSS-Image fix was proposed for ComfyUI yesterday, relating to the newly-added support for MOSS-image color and depth control inputs. https://github.com/Comfy-Org/ComfyUI/pull/17237#discussion_r4329744669 -> A new MOSS-Image Manga Editor. "Generate manga comics using an agent/skill, and make manual adjustments to the generated content." Looks like it could probably also work for western comic-book pages? https://github.com/Lotus0614/moss-manga-editor https://github-com.translate.goog/Lotus0614/moss-manga-editor?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp (English translation) -> ComfyUI nodes for Soda Storyboard... "import storyboard bundles / MOSS .JSON, preview, and MOSS-Image page workflows". Also good for comic-book page layout, by the look of it? https://github.com/cOkieeman/ComfyUI-Soda-Storyboard https://github-com.translate.goog/cOkieeman/ComfyUI-Soda-Storyboard?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp (English translation) -> A new MOSS-Image Base LoRA and LyCORIS trainer. No mention of LyCORIS in the translated readme, though, that I could see. https://github.com/waw1w1/moss_image_trainer https://github-com.translate.goog/waw1w1/moss_image_trainer?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp (English translation) -> MOSS-Image Edit, another try at making an Edit version and apparently supporting both Base and SFT. With a LoRA and two workflows (single, or up-to-four references). No explanation about the method by which it works. Huge *if* it works, but it doesn't for me - I can't even get it started because it has an unusual unified model + lora loader node. Even adding direct paths to the files couldn't get this loader node working. I'd be interested to hear if MOSS-Image Edit does work as advertised, and if a more user-friendly int8 + LoRA workflow has been made for it. *(Update: Kijai now has a request to add ComfyUI support - link below)* https://github.com/Chenkin-x/ComfyUI-Moss-Image-Edit/blob/main/README.en.md https://github.com/Comfy-Org/ComfyUI/pull/17296 -> A small collection of tested test-rendering prompts, plus tips useful for beginners. https://github.com/kaixin20190930/mossimage-prompts -> And finally, a new MOSS-Image-SFT-ROCmFPX conversion, for those with AMD graphics cards. https://huggingface.co/Harsha001/MOSS-Image-SFT-ROCmFPX ~ OLD POSTS ~ https://old.reddit.com/r/comfyui/comments/1xbw1jj/a_quick_mossimage_news_roundup_21st_october_2026/ https://old.reddit.com/r/comfyui/comments/1xarop9/a_quick_mossimage_news_roundup_20th_october_2026/ https://old.reddit.com/r/comfyui/comments/1xa3yk2/a_quick_mossimage_news_roundup_19th_october_2026/ https://old.reddit.com/r/comfyui/comments/1x97q81/a_quick_mossimage_news_roundup_18th_october_2026/
document fetched 2026-10-23 00:46 · canonical https://reddit.com/r/comfyui/comments/1xcm9wn/a_quick_mossimage_news_roundup_22nd_october_2026