import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("yehiaserag/anime-pencil-diffusion", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Anime-Pencil-Diffusion
A dreambooth finetune of stable diffusion 1.5 model that will output stuff in anime pencil concept drawing style.
Usage
Follow the directions under each version.
Anime-Pencil-Diffusion-V5
Trained for 400,000 steps, constant learning rate of 0.0000002 on 5000 images with 0 images for regularization.
Examples generated by the v5 model
Usage
Include animepencilconcept style in prompt to invoke the finetuned style.
Prompet comparison for V5
Anime-Pencil-Diffusion-V4
Trained for 160,000 steps, constant learning rate of 0.000001 on 526 images with 0 images for regularization and no text encoder training
Examples generated by the v4 model
Usage
Add the words anime pencil concept style anywhere in your prompt.
Anime-Pencil-Diffusion-V3
Trained for 12,000 steps, constant learning rate of 0.0000005 on 80 images with 1000 images of illustration style for regularization
Examples generated by the v3 model
Usage
Add the words anime pencil concept style anywhere in your prompt.
Anime-Pencil-Diffusion-V2
Trained for 4,000 steps, constant learning rate of 0.00000172 on 40 images with 1000 images of illustration style for regularization
Examples generated by the v2 model
Usage
Add the words anime pencil concept style anywhere in your prompt.
Anime-Pencil-Diffusion-V1
Trained on 2,400 steps, constant learning rate of 0.00000172 on 16 images with 1000 images of illustration style for regularization
Examples generated by the v1 model
Usage
Add the words anime pencil concept style anywhere in your prompt.
Socials
- Use the #AnimePencilConceptStyle so i can see the cool stuff you make!
- If you enjoy the model i'd appreciate a follow on twitter
- If you are feeling especially generous, you can sponsor me on paypal
- Created by Yehia Serag
NOTE: usage of this model implies accpetance of stable diffusion's CreativeML Open RAIL-M license
- Downloads last month
- 947