Skip to content

fix: nvbug_5398806 #6239

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
Jul 23, 2025
Merged

fix: nvbug_5398806 #6239

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
4 changes: 4 additions & 0 deletions tensorrt_llm/executor/result.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -228,6 +228,10 @@ def _handle_sequence(self,
output.logprobs = response_tensors.log_probs[src_idx]
# overcome some WAR in the cpp executor
if finish_reasons[src_idx] != tllm.FinishReason.CANCELLED:
if len(output.logprobs) > output.length:
# LlmResult holds a reference to LogProbStorage, which may be updated by the worker before the result is serialized.
# Therefore, we treat extra logprobs/logits as expected and only consume what's needed.
output.logprobs = output.logprobs[:output.length]
assert len(output.logprobs) == output.length
if response_tensors.generation_logits is not None:
output.generation_logits = response_tensors.generation_logits[
Expand Down