import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("jimmycarter/auraflow-4.8b", torch_dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]What is this
This is AuraFlow pruned to 18 layers from 36 layers, by excising the middle 18 layers. It will make mainly corrupt images until you finetune it for a few thousand steps, after which it should look good.
Usage
pipe = AuraFlowPipeline.from_pretrained("fal/AuraFlow")
del pipe.transformer
transformer = AuraFlowTransformer2DModel.from_pretrained(
"jimmycarter/auraflow-4.8b")
pipe.transformer = transformer
- Downloads last month
- 1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support