Sign In

Raehoshi illust XL

824
7k
251
Early Access
Verified:
SafeTensor
Type
Checkpoint Trained
Stats
521
4
Reviews
Published
Oct 12, 2024
Base Model
Illustrious
Training
Epochs: 15
Hash
AutoV2
88DF6E131F
Raelina's Avatar
Raelina

Raehoshi Illust XL

an enhanced iteration built upon the Illustrious XL model. It aims to elevate the visual style by addressing some of the limitations in the original, such as oversaturation and artifact noise. While these issues are not entirely eliminated, noticeable improvements have been made. The goal is to deliver a more polished, balanced output while staying true to the strengths of the base model.


Why Early Access?

Early access helps cover the cost of training and development. By purchasing early access, you're directly supporting the development of my models. If you'd like to support me further, you can also buy me a coffee on Ko-fi! Your support keeps the project going and ensures even better models in the future.


Positive prompt :

masterpiece, best quality, good quality

Negative prompt :

lowres, bad quality, worst quality, bad anatomy, sketch, jpeg artifacts, signature, watermark

Steps : 25-28

CFG : 5-6

Sampler : euler_a

Supported resolution :

1024 x 1024, 1152 x 896, 896 x 1152, 1216 x 832, 832 x 1216, 1344 x 768, 768 x 1344, 1536 x 640, 640 x 1536

Hires.fix Setting:

Training Details

The model was developed using a two-stage fine-tuning process. In Stage 1, new series and characters were introduced into the model. Stage 2 focused on fixing issues and enhancing the overall style for improved output.

Stage 1

  • Dataset : v1-31k, v2-37k, v3-34k, v4-60k

  • Hardware : 2x A100 80gb, v3 and v4-2x H100 80gb

  • Batch size : 32

  • Gradient accumulation steps : 2

  • Learning rate : 6e-6

  • Text encoder : 3e-6

  • Epoch : 15

Stage 2

  • Dataset : v1-2.5k, v2 and v3-2.3k, v4-2.5k

  • Hardware : 1x A100 80gb

  • Batch size : 48

  • Gradient accumulation steps : 1

  • Learning rate : 3e-6

  • Text encoder : disable

  • Epoch : 15

List of New Series/Characters Trained:

  • Zenless zone zero series character

  • Wuthering waves series character

  • Hololive from justice and dev is

  • HSR character from 2.2 - 3.2 version

  • Genshin character from 5.0 - 5.4 version

  • Tokidoki bosotto roshia-go de dereru tonari no alya-san series

  • Dandadan

  • Make heroine ga oo sugiru

For character prompts, please refer to the Danbooru site for accurate tags and references.

License

Fair AI Public License 1.0-SD

Special thanks to Joe for supporting my works