How to use from the
Use from the
Diffusers library
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline

# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Meina/MeinaPastel_V6", torch_dtype=torch.bfloat16, device_map="cuda")

prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]

MeinaPastel aims to make illustrations with a 2d feel to them with good light, shadows and details, making pastel or colorful images!

-- Recommendations of use:

  • Sampling method: DPM++ 2M Karass, 20 steps.
  • Upscaler: Latent(Nearest-Exact) at 15 steps and 0,55 Denoising in 2x.
  • Resolution: 512x768, 512x1024 , 768x512 , 1024x512 , 1536x512.
  • The VAE is baked-in.
  • Clip skip 2.

-- If you like the model and wants to support me in being able to spend more time improving it: -- You can do so by buying me a coffee at: https://ko-fi.com/meina ! ( it is not necessary but will be highly appreciated )

This model is a unet block merge of mostly MeinaMix and Colormixed, ultracolorv4 and a few others with minor block weight taken.

Downloads last month
84
Safetensors
Model size
0.9B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Space using Meina/MeinaPastel_V6 1