KleinBase4B_PBRStyle.safetensors

44 MB · SafeTensor

Requires membership

<stylized-pbr-flux2>

This model allows you to generate images in a stylized physically based rendering animation aesthetic, comparable to the kind seen in major animated films. It can generate new images as well as convert existing images into the style. It has a bit of a bias towards warm, natural light being cast from one side, so if you want different light conditions, it can help to specify the desired lighting; it also has a bias towards removing dark lipstick or removing atypical eye colorings, so I recommend specifying those in your prompt as well. In version 2.0, you simply need to include the trigger word "PBRSty1e" in your prompt and it will render in the style; this version supports both T2I and I2I workflows. In version 1.0, The primary way to utilize this model is to use the trigger phrase "make this PBRSty1e." This does have partial support for targeting specific subjects, so triggers like these may also work, depending on your source image: make the man PBRSty1e make the woman PBRSty1e Recommended Strength: 1.0-1.5 California AB 2013 Training Data Disclosure - This LoRA was fine-tuned using visual data consisting of still images derived from animated works, along with photographs, game screenshots, and AI-generated images used as inputs for synthetic transformation. The training data includes copyrighted material owned by third parties, including animation studios, game publishers, and individual artists, as well as derivative synthetic content generated from those inputs. No training data was licensed or purchased. This LoRA is provided for non-commercial use only under the terms of its distribution. - Version 1.0 Dataset: The dataset consists of 70 image pairs (140 images total). Each pair was created by taking an existing image, including animation stills, photographs, game screenshots, or AI-generated images, then generating an appropriate paired version of that image using AI, resulting a pair consisting of an image in the desired style as well as an image in the source style. Data was collected and generated between approximately 2024 and 2026. - Version 2.0 Dataset: The dataset consists of over one thousand images and contains both standalone image data as well as image pairs. This dataset consists of animation stills, photographs, game screenshots, and AI-generated images. - Image data was processed through standard resizing, cropping, normalization, and labeling steps. Synthetic images and video were included as part of the training dataset. - This model is intended for non-commercial, experimental, and educational use. Generated outputs may reflect copyrighted visual styles or themes associated with the underlying training data. Users are responsible for ensuring compliance with applicable copyright law, other intellectual property laws, and all other applicable laws.

SHA256

3027DEF194AD388089FB7DA9CC9511DE75B55BD842F48AD3B3A479EA3B9B839A

ModelTrends.ai Model ID

#26318

Sampler

Euler

Size

2548x960

Sampler

Euler

Size

2048x1024

Sampler

Euler

Size

2048x1024

Sampler

Euler

Size

2940x1024

Sampler

Euler

Size

2048x1024

Sampler

Euler

Size

1792x1152

Sampler

Euler

Size

2048x1024

Sampler

Euler

Size

1472x736