SBMM75/Krea-2-Raw_samples_Best_of

This dataset is a highly diverse set of high quality images generated with Krea 2 Raw.

NOTE: Raw is not intended for image generation, so do not use these images to judge the quality of the model. Raw is intended for training, as are the samples in this dataset as they can be used for regularization.

Possible uses

• —Regularization images for training models based on Krea 2 Raw
• —Quality testing

Data source

This dataset is derived from stablellama/FLUX.2-klein-base-9B_samples and was manually filtered to contain only images that have at least good quality, like correct anatomy.

The images were created in ComfyUI with the bf16 version of Krea 2 Raw. For each prompt were four images generated, all are (without any cherry picking) included in the corresponding dataset directories.

• —bf16 - published full model weights
• —1024x1024 pixels - native resolution for 1:1 aspect ratio images
• —batch size = 4
• —fixed seed (seed = 1)
• —workflow is the ComfyUI default for Krea 2
• —52 steps (as used by Krea at their Huggingface model documentation)
• —CGF = 3.5 (as used by Krea)
• —sampler: Euler
• —scheduler: simple

Data content

The data attempts to cover a very wide range of concepts. They are categorized into:

• —animals & creatures (Biological Diversity) (80 images)
• —architecture (60 images)
• —clothing (70 images)
• —composition (40 images)
• —DiffusionDB, originals ones and prompts but refined by a LLM (170 images)
• —dynamic poses (80 images)
• —hands (80 images)
• —handicraft (20 images)
• —humans (80 images)
• —interaction (40 images)
• —landscape (20 images)
• —logo, English (80 images)
• —noteable persons (17 images)
• —objects (100 images)
• —style (80 images)
• —Chinese text (100 images)
• —English text (100 images)

In total are 1217 different prompts and images. Contact prints for each category are at https://huggingface.co/datasets/stablellama/Krea-2-RawsamplesBestof/tree/main/contactprint

The images are stored in lossless compressed JPEG XL format.