A collection of styles that I like.
Show imagination and creativity. The style is not limited to the pictures from the preview.
You can find my guide to creating LoRa Styles, Early access and Exclusive models on my https://boosty.to/blacksnowskill or just support me 💜
https://t.me/blacksnowskill - news
Notice: This model is protected by copyright. Copying, distribution, merging with other models, or use on other websites without the explicit written permission of the author is prohibited.
Description
FAQ
Comments (6)
What's the difference between this and 1.0? The blurry effect and yellowish tint?
This version was trained in higher resolution and with improved descriptions.
@llikswonskcalb Oh btw, glad to see you're back. I just read about the whole banning thing. Edit: I thought you were an imposter stealing models at first hahah.
Does it work with A1111? It's somehow not showing up in my loRA list in the webui. Downloaded it twice, checked if it's in the loRA folder but won't show up. Am I missing something? X_X
me to
create a py script with the contents:
from safetensors.torch import load_file, save_file
input_path = "C:/Users/Username/Desktop/lora_project/DRKWB_modified.safetensors"
output_path = "C:/Users/Username/Desktop/lora_project/DRKWB_forge.safetensors"
state_dict = load_file(input_path)
metadata = {
"ss_version": "1",
"ss_network_module": "networks.lora",
"ss_network_alpha": "16",
"ss_network_dim": "32",
"ss_sd_model_name": "1761560.safetensors",
"ss_output_name": "DRKWB",
"ss_clip_skip": "2",
"ss_base_model_version": "sdxl_base_v1-0",
"modelspec.architecture": "stable-diffusion-xl-v1-base/lora"
}
save_file(state_dict, output_path, metadata=metadata)
print(f"LoRa was saved with Forge-compatible metadata: {output_path}")
_____________________________________________________
replace the path with the correct one and the names of the models, including the name in the metadata, run the script and it will create a new model file with the modified data that will work with Forge.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.