# A quick MOSS-Image news round-up - 24th October 2026 — retained record

> Retained reddit record from reddit:comfyui.

Canonical: https://highsignal.app/data/records/431aaf4e1cb67d66

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:comfyui)  published 2026-10-24 11:29 ingested 2026-10-25 02:39

[original source ↗](https://www.reddit.com/r/comfyui/comments/1xe69ab/a_quick_mossimage_news_roundup_24th_october_2026/)

## retained text



```
Another quick MOSS-Image news and goodies round-up, for those who may have missed some items. 

-> ComfyUI-Alt-Prompt. Making MOSS prompting even more old-school, by giving you... "A1111-style alternating and switching prompts". Working on MOSS-Image. Has example prompts and workflow.

https://github.com/silly2342/ComfyUI-Alt-Prompt

-> Following the success of adding Control references and Edit, now comes the first proof-of-concept attempt at a MOSS-Image LoRA that outputs clean .PNG cutouts with alpha channels. Working, and the user... "gets a transparent PNG straight out of the sampler: no chroma key, no background-removal model, and partial transparency where it belongs (glass, fairy wings, gems, lattices)." If you don't need semi-transparency for fairy-wings etc, then CivitAI also has a no-LoRA chroma-key based ComfyUI workflow for making cutout .PNGs with transparency.

https://huggingface.co/paralaif/MOSS-Image-Alpha-LoRA

https://civitai.com/models/3017655/moss-image-alpha-lora-transparent-png-proof-of-concept

https://civitai.com/models/3017440/moss-image-generation-with-alpha (no-LoRA chroma-key workflow)

-> An initial LoHA adapter for MOSS-Image Base... "trained to improve generation at 4mpx and higher resolutions." 

https://huggingface.co/comfyanonymous/MOSS_image_MAX_ULTRA_hires_lora_test

-> The important Spectrum for MOSS accelerator has a small fix today, affecting users who run workflows with *er_sde* instead of *euler*... "respect user horizon cap in ER-SDE adaptive".

https://github.com/Rkkss/ComfyUI-Spectrum-MOSS

-> Detailed training advice on character-LoRA likeness, along with settings.

https://old.reddit.com/r/malcolmrey/comments/1xdyjea/my_moss_image_base_character_lokr_training/
 
-> Comfyui-LoraQKV-Fix. "Make MOSS-Image's LoRA load correctly in ComfyUI: automatically handles the differences between separating and merging q/k/v, as well as the differences in key naming among various LoRA trainers."

https://github.com/storyAura/Comfyui-LoraQKV-Fix

https://github-com.translate.goog/storyAura/Comfyui-LoraQKV-Fix?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en&_x_tr_pto=wapp (English translation)

-> MOSS Afterlight. A small LoRA that consistently... "adds that warm, cinematic feeling you get when the sun is low and everything suddenly looks better. It brings in soft golden light, natural glow, creamy background blur, and a bit of atmosphere without completely changing the image." Appears to be a port from the popular Krea2 Afterlight LoRA. No trigger.

https://civitai.com/models/3016917/moss-afterlight

-> Nu-Moss, a flat-ish semi-toon illustration LoRA for MOSS-Image. No trigger.

https://civitai.com/models/3015064/nu-moss-style-a-semi-deformed-illustration

-> A couple of MOSS-Image Lycoris I missed. Pleasing flat-ish lineart styles meant for anime, but which look like they might also be useful in a western comics workflow - provided they don't force the generation towards 'huge eyes, small noses, young female' anime.

https://civitai.com/models/3010674/nizuma-style-moss-base (working on Base only)

https://civitai.com/models/3010642/souichi60-style-moss-base (also works on SFT with trigger *@souichi60 lineart*)

-> And finally, a YouTube tutorial on controlling MOSS-image using three approaches: .JSON prompts with co-ordinates; BBox control via visual bounding boxes; and Colour-Depth controlnet references. Along the way he also finds that a Canny reference can work too.

https://www.youtube.com/watch?v=e3gsjI-hxCY

~ OLD POSTS ~

https://old.reddit.com/r/comfyui/comments/1xdrwcu/a_quick_mossimage_news_roundup_23rd_october_2026/ (bonus LoRA edition)

https://old.reddit.com/r/comfyui/comments/1xd43i2/a_quick_mossimage_news_roundup_23rd_october_2026/

https://old.reddit.com/r/comfyui/comments/1xcm9wn/a_quick_mossimage_news_roundup_22nd_october_2026/

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 2026-10-25 00:36 · canonical https://reddit.com/r/comfyui/comments/1xe69ab/a_quick_mossimage_news_roundup_24th_october_2026
