Sign In

NewArtModel4

167
556
3
Verified:
SafeTensor
Type
Checkpoint Trained
Stats
556
Reviews
Published
Feb 6, 2023
Base Model
SD 1.5
Training
Steps: 3,500
Trigger Words
m_style
Hash
AutoV2
7F5E6E5863
Nerfgun3's Avatar
Nerfgun3

Why NewArtModel?

The model itself is mainly based on Waifu-Diffusion 1.3 and was finetuned using a dataset of my own artwork and collected artwork of others.

I hope you enjoy the model. If you have any questions, you can ask me anything via Discord: "Nerfgun3#7508"

Use Cases

The model itself is very good in itself and produces high quality images from landscapes to portraits to anime. However, it shines the most when merged with other models.

Here are a few examples:

  1. Generally all models from andite

  2. Kenshi Model from Luna

Token

This model was trained on a token m_style like the Dreambooth models, but as can be seen in the sample prompts, it is not needed, and since merging is one of its strengths, it is not essential.