import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("TheLastBen/froggy-style", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Froggy Style V1.5
V1.5 Model by TheLastBen
This model is trained on 11 Midjourney images 512x512, 1300 steps and 300 steps text_encoder (30% because the total steps is low, normally 15%)
Prompts to start with :
ttdddd , __________, movie, ultra high quality render, high quality graphical details, 8k, volumetric lighting, micro details, (cinematic)
Negative : bad, low-quality, 3d, game
The prompt also can be as simple as the instance name : ttdddd and you will still get great results.
You can also train your own concepts and upload them to the library by using fast-DreamBooth.
Test the concept via A1111 Colab :fast-stable-diffusion-A1111
Sample pictures of this concept:
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
.png)
- Downloads last month
- 49