import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("fill-in-base-model", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("logo-wizard/logo-diffusion-checkpoint")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]INFO
This is checkpoint based on stabilityai/stable-diffusion-2-1 and logo-wizard/logo-diffusion. The weights were fine-tuned on the logo-wizard/modern-logo-dataset dataset. You can find some example images in the following.
Best practices
We recommend using this model with the following prompt template:
positive: f"a logo of {company industry}, {some objects}, {colors}, modern, minimalism, vector art, 2d, best quality, centered"
negative: "low quality, worst quality, bad composition, extra digit, fewer digits, text, inscription, watermark, label, asymmetric"
Some other recommendations:
num_inference_steps = 30
guidance_scale = 7.5
height = 768
width = 768
scheduler = diffusers.EulerAncestralDiscreteScheduler (used by default)
- Downloads last month
- 24