Hey, tried to train my Girls und Panzer dataset consisting of Megami Magazine artworks of the franchise on Moss.Works the best together with Lodestone's 8step dmd lora: <a target="_blank" rel="ugc" href="https://huggingface.co/lodestones/MOSS-Image-SFT-8-steps-dmd/blob/main/lora_latest.safetensors">https://huggingface.co/lodestones/MOSS-Image-SFT-8-steps-dmd/blob/main/lora_latest.safetensors</a>Recommended settings: CFG: 2 (yes, not 1. You really want the negative prompt to work for this model)Steps: 8Euler + Simple or whatever makes senseTriggerword: Girls und Panzeror Girls und Panzer anime illustrationLoRa Strength: 0.8

Girls und Panzer MOSS Style

Trained on MOSS Image Base and tested on Base, not sft.

Files with SHA256: 244b56c4966b462bfe65a5551246c3dd79d42567be455c7b0ac30bd89050e2ce