Dataset Viewer
Auto-converted to Parquet Duplicate
Search is not available for this dataset
image
image
YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/datasets-cards)

CKN12000 artist style test

https://huggingface.co/ChenkinNoob

The prompt parameters used for testing is:

1girl, sangonomiya kokomi,genshin impact, sky,smile,full body,bowtie,pink hair,bow,simple background,smile,loli,wariza, masterpiece,newest,absurdres,safe,

Negative prompt: furry,(nsfw:1.2), bad quality,poor quality,doll,disfigured,jpg,toy,bad anatomy,missing limbs,missing fingers,3d,cgi,text,watermark,bad anatomy,bad proportions,extra limbs,extra digit,extra legs,extra legs and arms,disfigured,missing arms,too many fingers,fused fingers,missing fingers,unclear eyes,watermark,username, Steps: 20, Sampler: Euler a, Schedule type: Simple, CFG scale: 5, Seed: 1145141919, Size: 1024x1536, Model hash: c94407a670, Model: ckn0.3-ep2, Version: f2.0.1v1.10.1-previous-636-gb835f24a

Note:
  1. This content is for reference only and should not be used as evidence that "the model contains a specific style."
  2. It is strictly prohibited to open this document with WPS/Office. It is recommended to open it locally using Tencent Docs (to avoid memory issues). If Tencent Docs cannot open it, please try the paginated version.
  3. Under no circumstances should you click the Save/Auto-save button. If a save prompt appears when closing, click "No" to avoid a secondary save, otherwise the document size may expand to 55GB.
We do not recommend using this document for the following groups of people, and we regret any inconvenience:
  1. Individuals who cannot tolerate any deviation from the original artistic style/character traits, even in the most minute details.
  2. Individuals facing application scenarios with extremely high requirements for the accuracy of reproducing artistic styles/character traits.
  3. Individuals who cannot accept the potential randomness inherent in AI-generated images based on the Stable Diffusion algorithm.
  4. Individuals who cannot accept the use of fully automated processes for image generation and tuning, or who believe that maintaining the integrity of the original art necessitates completely manual operation for image generation.
  5. Individuals who find the generated image content offensive to their values, or who believe that AI image generation and automated training technologies are disrespectful to them.
The images generated without any style prompts are as shown in the figure below.

Downloads last month
209