2023/12/24 update

New version is retrained using the latest technology to make generated images more similar to the dataset. Merry Christmas ! XD

2023/6/26 update

Update dataset. Retrain the model using new craft. (The latest version is LoRA model).

2023/4/15 update

Update dataset and upload lycoris version. Adjust trigger tag to soyincapazz, username of this artist. The preview images of the new model are generated using 'Hires Fix' and upscaled with 'SD Upscale'.

2023/4/1 upate

Fix the compatibility problem of non-NAI-based checkpoints. Now arbitrary anime model with NAI's VAE or kl-f8-anime2 VAE can also generate good results using this LoRA, theoretically.

2023/3/1 update

Reduce loss to 0.09. 0.6 and 0.7 weight should be enough.

Introduction

A LORA model of Tsuki no i-min style. The 784MB VAEs are 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.