Impermanence_Anime Fantasy Girl F.1

Introduction to the selected LORA model: This model is trained based on the F.1-dev-fp8 base model, with an anime fantasy style.

Recommendation: Strongly recommend using the Flux-generated image workflow, 3 computational power for one image, what are you waiting for?

Workflow link: https://www.liblib.art/modelinfo/9c81aa8555d2425490f7ecdf028f9abd?from=personal_page

Generation suggestions: LORABase algorithm model - runnable online at LiBu,

Sampling method: Euler

LORA weight: Recommended 0.8-1.0

Recommended size: This model is trained on a 1024*1536 training image set.

Number of iterations: 15-30, tested to produce decent images even at 10 steps

Prompt word correlation coefficient: 3.5.

HD restoration: R-ESRGAN_4x+ Anime6B

Creative overview: Trying to train a new anime fantasy-style LORA model using the newly launched FLUX. Unexpected surprises of anime characters under the FLUX large model? Recommend using the F.1 model of the image workflow to save a lot of computational power.

Trigger words: Anime Girl