Skip to content

Update regional_aot.py (profile links) #3559

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 2 commits into from
Sep 12, 2025
Merged

Update regional_aot.py (profile links) #3559

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
db53c7a
Update regional_aot.py
sayakpaul Sep 12, 2025
b16f954
Update regional_aot.py
sayakpaul Sep 12, 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
6 changes: 4 additions & 2 deletions recipes_source/regional_aot.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,7 @@
Reducing AoT cold start compilation time with regional compilation
============================================================================

**Author:** `Sayak Paul <https://github.com/sayakpaul>`_, `Charles Bensimon <https://github.com/cbensimon>`_, `Angela Yi <https://github.com/angelayi>`_
**Author:** `Sayak Paul <https://huggingface.co/sayakpaul>`_, `Charles Bensimon <https://huggingface.co/cbensimon>`_, `Angela Yi <https://github.com/angelayi>`_

In the `regional compilation recipe <https://docs.pytorch.org/tutorials/recipes/regional_compilation.html>`__, we showed
how to reduce cold start compilation times while retaining (almost) full compilation benefits. This was demonstrated for
Expand Down Expand Up @@ -236,4 +236,6 @@ def aot_compile_load_model(regional=False) -> torch.nn.Module:
#
# This recipe shows how to control the cold start time when compiling your
# model ahead-of-time. This becomes effective when your model has repeated
# blocks, which is typically seen in large generative models.
# blocks, which is typically seen in large generative models. We used this
# recipe on various models to speed up real-time performance. Learn more
# `here <https://huggingface.co/blog/zerogpu-aoti>`__.