SanaSprintPipeline: Intermediate timesteps for SCM is not supported when num_inference_steps != 2

[nitinmukesh]

Describe the bug

Documentation says 1-4 inference steps, but using 4 steps throws error

achieving high-quality generation with 1-4 inference steps

https://github.com/huggingface/diffusers/blob/main/src/diffusers/pipelines/sana/pipeline_sana_sprint.py#L413

if intermediate_timesteps is not None and num_inference_steps != 2:
            raise ValueError("Intermediate timesteps for SCM is not supported when num_inference_steps != 2.")

Reproduction

from diffusers import SanaSprintPipeline
import torch

pipeline = SanaSprintPipeline.from_pretrained(
    "Efficient-Large-Model/Sana_Sprint_1.6B_1024px_diffusers",
    torch_dtype=torch.bfloat16
)
pipeline.to("cuda:0")
prompt = "a tiny astronaut hatching from an egg on the moon"

image = pipeline(prompt=prompt, num_inference_steps=4).images[0]
image.save("test_out.png")

Logs

(venv) C:\aiOWN\diffuser_webui>python Sana_Sprint.py
Loading checkpoint shards: 100%|████████████████████████████████████| 2/2 [00:00<00:00,  3.07it/s]
Loading pipeline components...: 100%|███████████████████████████████| 5/5 [00:05<00:00,  1.08s/it]
Traceback (most recent call last):
  File "C:\aiOWN\diffuser_webui\Sana_Sprint.py", line 14, in <module>
    image = pipeline(prompt=prompt, num_inference_steps=4).images[0]
  File "C:\aiOWN\diffuser_webui\venv\lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
  File "C:\aiOWN\diffuser_webui\venv\lib\site-packages\diffusers\pipelines\sana\pipeline_sana_sprint.py", line 725, in __call__
    self.check_inputs(
  File "C:\aiOWN\diffuser_webui\venv\lib\site-packages\diffusers\pipelines\sana\pipeline_sana_sprint.py", line 414, in check_inputs
    raise ValueError("Intermediate timesteps for SCM is not supported when num_inference_steps != 2.")
ValueError: Intermediate timesteps for SCM is not supported when num_inference_steps != 2.

System Info

Build from source

Who can help?

No response

[nitinmukesh]

Solved.