Skip to content

Fix Context Parallelism doc #12631

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 2 commits into
base: main
Choose a base branch
from
Open

Fix Context Parallelism doc #12631

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f2ea6b4
Fix Context Parallelism doc
a120092009 Nov 11, 2025
c83484a
Merge branch 'main' into fix_cp_doc
sayakpaul Nov 11, 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
4 changes: 2 additions & 2 deletions docs/source/en/training/distributed_inference.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -253,7 +253,7 @@ try:
device = torch.device("cuda", rank % torch.cuda.device_count())
torch.cuda.set_device(device)

transformer = AutoModel.from_pretrained("Qwen/Qwen-Image", subfolder="transformer", torch_dtype=torch.bfloat16, parallel_config=ContextParallelConfig(ring_degree=2))
transformer = AutoModel.from_pretrained("Qwen/Qwen-Image", subfolder="transformer", torch_dtype=torch.bfloat16, parallel_config=ContextParallelConfig(ring_degree=2), device_map="cuda")
Copy link
Member

@sayakpaul sayakpaul Nov 11, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

There's device_map="cuda" just in the pipeline loading code. Is that not enough?

Copy link
Contributor Author

@a120092009 a120092009 Nov 11, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

yes, error will occur without device_map

image

Copy link
Member

@sayakpaul sayakpaul Nov 11, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh I see. Good catch.

Copy link
Contributor Author

@a120092009 a120092009 Nov 13, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So is it a desired behavior?

Copy link
Member

@sayakpaul sayakpaul Nov 13, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah it should be because CP relies on a compatible device.

pipeline = QwenImagePipeline.from_pretrained("Qwen/Qwen-Image", transformer=transformer, torch_dtype=torch.bfloat16, device_map="cuda")
pipeline.transformer.set_attention_backend("flash")

Expand Down Expand Up @@ -289,4 +289,4 @@ Pass the [`ContextParallelConfig`] to [`~ModelMixin.enable_parallelism`].

```py
pipeline.transformer.enable_parallelism(config=ContextParallelConfig(ulysses_degree=2))
```
```