Skip to content

generate: drop stale num_return_sequences warning on continuous batching path #45582

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
remi-or merged 4 commits into from
Apr 24, 2026
Merged

generate: drop stale num_return_sequences warning on continuous batching path #45582

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
ad4c4f8
generate: drop stale num_return_sequences warning on continuous batch…
joaquinhuigomez Apr 22, 2026
bb18772
Apply repo consistency fixes
github-actions[bot] Apr 23, 2026
4888862
Merge branch 'main' into fix/paged-num-return-sequences-warning
remi-or Apr 23, 2026
89af6a5
Merge branch 'main' into fix/paged-num-return-sequences-warning
remi-or Apr 24, 2026
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: 2 additions & 6 deletions src/transformers/generation/utils.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2303,13 +2303,9 @@ def generate(
# others are ignored
if synced_gpus is not None:
logger.warning(f"synced_gpus is not ignored for continuous batching. Got {synced_gpus = }")
num_return_sequences = kwargs.get("num_return_sequences", 1)
num_beams = kwargs.get("num_beams", 1)
if num_return_sequences > 1 or num_beams > 1: # FIXME: remove this once CB supports it (which is planned)
logger.warning(
f"num_return_sequences and num_beams are not supported for continuous batching yet. "
f"Got {num_return_sequences = } and {num_beams = }. "
)
if num_beams > 1: # FIXME: remove this once CB supports num_beams (which is planned)
logger.warning(f"num_beams is not supported for continuous batching yet. Got {num_beams = }. ")

# switch to CB
outputs = self.generate_batch(
Expand Down
Loading