How to use stabilityai/stable-diffusion-3.5-medium_amdgpu with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("stabilityai/stable-diffusion-3.5-medium_amdgpu", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0]
Is there a reason the text_encoder hidden_states is suffixed with 1, this makes it incompatible with huggingface optimum. Is this onnx model only supported by Amuse GUI out of the box?
hidden_states
1
· Sign up or log in to comment