OMCCOM/Holo Spice And Wolf (MERCHANT MEETS THE WISE WOLF)

I've retrained the model for MOSS-Image; it seems the base model wasn't properly trained for Holo. I haven't finalized the training parameters yet, but it can generate images reasonably well, so I'm releasing it. Try specifying a weight of 0.75; 1.0 is too strong when using the MOSS-Image concept, which isn't included in this LoRA. I've tried generation parameters of step=50 CFG=3.0, and it seems possible to generate 4MP images without using an upscaler. Currently, it's trained with booru-compatible tags, but to improve accuracy, I might need to incorporate natural language that specifies the

<iframe src="https://silytics.si/embed/item/d5f80844-709a-4267-a238-c1aa0464d2e1" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Civitai and links to the original. Information may be incomplete or out of date. Last refreshed 23 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.