Skip to content

[CB] set new_tokens to to max value given the constraints #516

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
Oct 10, 2025
Merged

[CB] set new_tokens to to max value given the constraints #516

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
7 changes: 6 additions & 1 deletion vllm_spyre/platform.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -505,7 +505,12 @@ def get_max_output_tokens(self, prompt_len: int) -> int:
"""Return the size of biggest ```new_tokens``` of the \
warmup shapes that fits the prompt length"""
if self._warmup_shapes is None:
return sys.maxsize
# ceil division to pad to next block boundary
padded_prompt_len = math.ceil(
prompt_len / self._block_size) * self._block_size
max_new_tokens = (self._config.scheduler_config.max_model_len -
padded_prompt_len)
return max_new_tokens

max_new_tokens = 1
for shape in self._warmup_shapes:
Expand Down