import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-xl-base-1.0", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("TheLastBen/THE_FISH")
prompt = "daiton"
image = pipe(prompt).images[0]THE FISH
SDXL LoRA by TheLastBen
Prompts to start with :
(daiton:1.2) (giant:1.15) demon, inside a dark creepy abandoned mine at (night:1.1), (dolly zoom:1.1) , flash photography, by william eggleston, closeup, cinematic, still frame, 4k
ComfyUI is recommended for best quality,
Use the sample images to load the settings onto the UI.
Trained using https://github.com/TheLastBen/fast-stable-diffusion SDXL trainer.
Sample pictures:
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
- Downloads last month
- 44
Model tree for TheLastBen/THE_FISH
Base model
stabilityai/stable-diffusion-xl-base-1.0