Skip to content

Fix: Normalize batch inputs to 5D tensors for Qwen-Image-Edit #12698

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

Open
wants to merge 7 commits into
base: main
Choose a base branch
from
Open

Fix: Normalize batch inputs to 5D tensors for Qwen-Image-Edit #12698

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
0c49735
Fix: Normalize batch inputs to 5D tensors for Qwen-Image-Edit
akshan-main Nov 22, 2025
1b85230
Fix: Handle variable sequence lengths in batch inference via padding
akshan-main Nov 22, 2025
da78501
Fix: Add routing logic for 1-to-1 batch inference and tokenizer padding
akshan-main Nov 22, 2025
2b05da8
Fix: Robust batch inference with padding and 1-to-1 stacking
akshan-main Nov 22, 2025
f84b079
Qwen-Image-Edit robust batch inference with padding
akshan-main Nov 22, 2025
22dd1cb
Fix Qwen batching: final logic for input resizing and robust inference
akshan-main Nov 24, 2025
474b0e9
Merge branch 'main' into fix-qwen-image-batching
akshan-main Nov 24, 2025
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
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/controlnet/pipeline_controlnet_sd_xl_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,8 +131,7 @@

>>> prompt = "A robot, 4k photo"
>>> image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... ).resize((1024, 1024))
>>> controlnet_conditioning_scale = 0.5 # recommended for good generalization
>>> depth_image = get_depth_map(image)
Expand Down
4 changes: 1 addition & 3 deletions src/diffusers/pipelines/cosmos/pipeline_cosmos_video2world.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,9 +88,7 @@ def __init__(self, *args, **kwargs):
>>> prompt = "The video depicts a winding mountain road covered in snow, with a single vehicle traveling along it. The road is flanked by steep, rocky cliffs and sparse vegetation. The landscape is characterized by rugged terrain and a river visible in the distance. The scene captures the solitude and beauty of a winter drive through a mountainous region."
>>> video = load_video(
... "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/cosmos/cosmos-video2world-input-vid.mp4"
... )[
... :21
... ] # This example uses only the first 21 frames
... )[:21] # This example uses only the first 21 frames
>>> video = pipe(video=video, prompt=prompt).frames[0]
>>> export_to_video(video, "output.mp4", fps=30)
Expand Down
2 changes: 1 addition & 1 deletion src/diffusers/pipelines/kandinsky/pipeline_kandinsky_combined.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,7 @@
negative_prompt = "low quality, bad quality"

original_image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/cat.png"
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
)

mask = np.zeros((768, 768), dtype=np.float32)
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky/pipeline_kandinsky_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,8 +60,7 @@
>>> pipe.to("cuda")

>>> init_image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/frog.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/frog.png"
... )

>>> image = pipe(
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky/pipeline_kandinsky_inpaint.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -66,8 +66,7 @@
>>> pipe.to("cuda")
>>> init_image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )
>>> mask = np.zeros((768, 768), dtype=np.float32)
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky/pipeline_kandinsky_prior.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -88,8 +88,7 @@
>>> pipe_prior.to("cuda")

>>> img1 = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )

>>> img2 = load_image(
Expand Down
2 changes: 1 addition & 1 deletion src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_combined.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,7 +92,7 @@
negative_prompt = "low quality, bad quality"
original_image = load_image(
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main" "/kandinsky/cat.png"
"https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
)
mask = np.zeros((768, 768), dtype=np.float32)
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -71,8 +71,7 @@


>>> img = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... ).resize((768, 768))

>>> hint = make_hint(img, depth_estimator).unsqueeze(0).half().to("cuda")
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_controlnet_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,8 +72,7 @@
>>> pipe = pipe.to("cuda")

>>> img = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... ).resize((768, 768))


Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,8 +56,7 @@
>>> pipe.to("cuda")

>>> init_image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/frog.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/frog.png"
... )

>>> image = pipe(
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_inpainting.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,8 +62,7 @@
>>> pipe.to("cuda")

>>> init_image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )

>>> mask = np.zeros((768, 768), dtype=np.float32)
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_prior.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,8 +60,7 @@
... )
>>> pipe_prior.to("cuda")
>>> img1 = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )
>>> img2 = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
Expand Down
6 changes: 2 additions & 4 deletions src/diffusers/pipelines/kandinsky2_2/pipeline_kandinsky2_2_prior_emb2emb.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,8 +35,7 @@
>>> prompt = "red cat, 4k photo"
>>> img = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )
>>> image_emb, nagative_image_emb = pipe_prior(prompt, image=img, strength=0.2).to_tuple()
Expand Down Expand Up @@ -73,8 +72,7 @@
>>> pipe_prior.to("cuda")
>>> img1 = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... )
>>> img2 = load_image(
Expand Down
3 changes: 1 addition & 2 deletions src/diffusers/pipelines/pag/pipeline_pag_controlnet_sd_xl_img2img.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,8 +132,7 @@

>>> prompt = "A robot, 4k photo"
>>> image = load_image(
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main"
... "/kandinsky/cat.png"
... "https://huggingface.co/datasets/hf-internal-testing/diffusers-images/resolve/main/kandinsky/cat.png"
... ).resize((1024, 1024))
>>> controlnet_conditioning_scale = 0.5 # recommended for good generalization
>>> depth_image = get_depth_map(image)
Expand Down
Loading