import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("John6666/blue-pencil-flux1-v001-fp8-flux", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Original model is here.
Notice
This is an experimental conversion in Spaces using a homebrew script. serverless Inference API does not currently support torch float8_e4m3fn, so it does not work. There are still many bugs around the FLUX.1 conversion part of Diffusers, and I have not been able to confirm if the conversion is working properly. Please consider this as a test run only.
- Downloads last month
- 3
Model tree for John6666/blue-pencil-flux1-v001-fp8-flux
Base model
bluepen5805/blue_pencil-flux1