import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("SeanScripts/pyramid-flow-sd3-bf16", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Converted to bfloat16 from rain1011/pyramid-flow-sd3. Use the text encoders and tokenizers from that repo (or from SD3), no point reuploading them over and over unchanged.
Inference code is available here: github.com/jy0205/Pyramid-Flow.
Both 384p and 768p work on 24 GB VRAM. For 16 steps (5 second video), 384p takes a little over a minute on a 3090, and 768p takes about 7 minutes. For 31 steps (10 second video), 384p took about 10 minutes.
I highly recommend using cpu_offloading=True when generating, unless you have more than 24 GB VRAM.
- Downloads last month
- -
Model tree for SeanScripts/pyramid-flow-sd3-bf16
Base model
stabilityai/stable-diffusion-3-medium Finetuned
rain1011/pyramid-flow-sd3