Hugging Face's logo

kayfahaarukku
/
irAsu-1.0

Text-to-Image
Diffusers
Safetensors
English
StableDiffusionXLPipeline
stable-diffusion
stable-diffusion-xl
Inference Endpoints
Model card Files Community
irAsu-1.0 / README.md
kayfahaarukku's picture
kayfahaarukku
Update README.md
524e369 verified
preview code
|
raw
history blame contribute delete
5.52 kB
metadata
license: other
license_name: faipl
license_link: https://freedevproject.org/faipl-1.0-sd
language:
  - en
tags:
  - text-to-image
  - stable-diffusion
  - safetensors
  - stable-diffusion-xl
base_model: OnomaAIResearch/Illustrious-xl-early-release-v0
widget:
  - text: >-
      1girl, green hair, sweater, looking at viewer, upper body, beanie,
      outdoors, night, turtleneck, masterpiece, best quality
    parameter:
      negative_prompt: >-
        nsfw, lowres, bad anatomy, bad hands, text, error, missing fingers,
        extra digit, fewer digits, cropped, worst quality, low quality, normal
        quality, jpeg artifacts, signature, watermark, username, blurry, artist
        name
    example_title: 1girl

irAsu v1.0

sample1
sample2

irAsu is a second-stage finetune of Illustrious XL 0.1, which fixes the coloring, anatomy, and background of the images generated by the model.

Standard Prompting Guidelines

The model is finetuned from Illustrious XL v0.1, which is trained with danbooru tags. However, there is a little bit changes on dataset captioning, therefore there is some different default prompt used:

Default prompt: 

1girl/1boy, character name, from what series, everything else in any order, best quality, amazing quality, very aesthetic.

Note: The quality tag masterpiece has been replaced with best quality due to reports that it often caused unwanted side effects.

Default negative prompt: 

lowres, (bad), text, error, fewer, extra, missing, worst quality, jpeg artifacts, low quality, watermark, unfinished, displeasing, oldest, early, chromatic aberration, signature, extra digits, artistic error, username, scan, [abstract],

Default configuration:

Default configuration: Euler a with around 25-30 steps, CFG 3.5-5, and ENSD set to 31337. Sweet spot is around 28 steps and CFG 4.

Training Configurations

Pretraining: N/A

Finetuning:

  • Dataset size: 3,140 images

  • GPU: 1xA100

  • Optimizer: AdaFactor

  • Unet Learning Rate: 3e-6

  • Text Encoder Learning Rate: - (Train TE set to False)

  • Batch Size: 16

  • Gradient Accumulation: 3

  • Warmup steps: 5%

  • Min SNR Gamma: 5

  • Epoch: 10 (epoch 9 is used)

  • Noise Offset: 0.0357

  • Random Cropping: True

  • Loss: Huber

  • Huber Schedule: SNR

  • Huber C: 0.1

License

irAsu v1.0 falls under the Fair AI Public License 1.0-SD license.