Shirakami Fubuki – A lora that someone already made but I still made one more :v
Trigger Words: photo of shirakami fubuki
Why is it so small in size?
-
I’m not sure why, as I used to export Lora from a fine-tuned model, the result is a tiny Lora file. Also, I don’t know how to train a Lora correctly. I used Dreambooth to train a new model and then exported Lora from the resulting model.
I’m interested in your Lora, please tell me how to use it!
-
Of course, I’m happy to share. First, download the model using the download button or if you using Colab:
!wget https://civitai.com/api/download/models/19244 -O /content/stable-diffusion-webui/models/Lora/shirakami_fubuki.safetensors
Or:
import requests url = https://civitai.com/api/download/models/19244 save_path = /content/stable-diffusion-webui/models/Lora/shirakami_fubuki.safetensors response = requests.get(url) with open(save_path, 'wb') as f: f.write(response.content)
-
You need to add the Lora tag “lora:shirakami_fubuki:1”, and add some tags like “fox ears, white hair”, finally, you can have a character similar to the picture. And of course, don’t forget “photo of shirakami fubuki”.
-
The images above were created using my Lora combined with the CetusMix and Grapefruit models.
-
Secret: If you accidentally add tags like “white kimono, umbrella, blue eyes, long hair”, you will have kawakaze =)))
Training:
-
Dataset: Trained on 50 images and 50 reinforcement images (other AI-generated images)
-
Base model: anime-base-sd-v15-v20LTS
-
Schedule: DDIM.
-
Optimizer: AdamW.
-
Precision: BF16.
-
Hardware: Google Colaboratory Pro – NVIDIA A100 40GB VRAM.
By Ojimi and SimpleThread