import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("hf-internal-testing/tiny-qwenimage21-turbo-modular", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]Tiny random-weight Qwen-Image 2.1 Turbo modular pipeline (QwenImage21TurboAutoBlocks) for diffusers CI tests.
Built from hf-internal-testing/tiny-qwenimage21-modular with the Turbo checkpoint's configs: sample_sigmas from
Qwen/Qwen-Image-2.1-Turbo in modular_model_index.json, and use_dynamic_shifting=false, shift_terminal=null in
the scheduler config.
- Downloads last month
- -