Skip to content

Fix num_images_per_prompt>1 with Skip Guidance Layers in StableDiffusion3Pipeline #10086

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 2, 2024
Merged

Fix num_images_per_prompt>1 with Skip Guidance Layers in StableDiffusion3Pipeline #10086

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 3 additions & 5 deletions src/diffusers/pipelines/stable_diffusion_3/pipeline_stable_diffusion_3.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -907,11 +907,7 @@ def __call__(
continue

# expand the latents if we are doing classifier free guidance
latent_model_input = (
torch.cat([latents] * 2)
if self.do_classifier_free_guidance and skip_guidance_layers is None
else latents
)
latent_model_input = torch.cat([latents] * 2) if self.do_classifier_free_guidance else latents
# broadcast to batch dimension in a way that's compatible with ONNX/Core ML
timestep = t.expand(latent_model_input.shape[0])

Expand All @@ -935,6 +931,8 @@ def __call__(
else False
)
if skip_guidance_layers is not None and should_skip_layers:
timestep = t.expand(latents.shape[0])
latent_model_input = latents
noise_pred_skip_layers = self.transformer(
hidden_states=latent_model_input,
timestep=timestep,
Expand Down
Loading