2023/10/21 update

Update dataset. Retrain the model using new craft. Using CCIP metric to measure model performance. The accuracy of this version under CCIP metric is about 97%.

2023/4/9 update

Update dataset. Upload lycoris version. v1.0 version is conventional LoRA model. The preview images are generated using 1.6 version (original) YesMix.

Introduction

A conventional LORA model of musashi (azur lane) trained using 160 highres images, including nsfw. The NAI's VAE is recommended. 0.6~0.8 weights are good.

Cautions (Must Read)

• Some preview images are generated using hires fix. Thus, if you wanna generate the preview images, pls download the original preview images and analyze them using 'PNG INFO' in webui to get all generating information.

• The maximum of generating resolution should not exceed 1024 theoretically, otherwise extra/deformed head/limbs will be waiting for you :). Try to use 'hires fix' to prevent this problem and help to generate higher-resolution images.

• For 784mb VAEs (NovelAI, Anythingv3, Orangemix, Counterfeit) users, make sure add '--no-half-vae' in command line of webui to prevent generating black images. See voldy's log for the details of edit method. For high-performance GPU, make sure 'medvram' and 'lowvram' mode are disabled.