import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("fbobe3/acestep-v15-xl-turbo-diffusers-mirror", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]ACE-Step v1.5 XL-Turbo (diffusers) — mirror
This is a verbatim mirror kept for reliability/reproducibility of the RemixFlow project. All weights and credit belong to the original authors.
Original model
- Model:
ACE-Step/acestep-v15-xl-turbo-diffusers - Project: ACE-Step 1.5 by the ACE-Step Team (ACE Studio & StepFun)
- License: Apache-2.0 (see the original repository)
Please prefer the original model; use this mirror only as a fallback.
Used by
RemixFlow — an AI music evolution
platform. Its ace-step backend drives this checkpoint via the diffusers
AceStepPipeline (SDEdit audio-to-audio) to generate identity-preserving
variations and endless "Living Songs".
from diffusers import AceStepPipeline
import torch
pipe = AceStepPipeline.from_pretrained(
"fbobe3/acestep-v15-xl-turbo-diffusers-mirror", torch_dtype=torch.bfloat16
).to("cuda")
- Downloads last month
- 11