import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Ryzan/fantasy-diffusion-v0", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]fantasy-diffusion-v0 diffusion model trained by Ryzan with DreamBooth
fantasy-diffusion-v1 is already out
v1 has better rendering than v0
see v1 at: https://huggingface.co/Ryzan/fantasy-diffusion-v1
v0 is currently a test model
This model is trained on 400 images of semi-realistic fantasy art
This model does not use any finetuning techniques such as face restoration or in-painting as of yet
Sample pictures of this concept:
| Prompt: fantastyle foxgirl | Prompt: fantastyle foxgirl |
|---|---|
 |
.jpg) |
| Prompt: fantastyle knight | Prompt: fantastyle knight |
|---|---|
 |
.jpg) |
| Prompt: fantastyle mage | Prompt: fantastyle mage |
|---|---|
 |
.jpg) |
- Downloads last month
- 10