Skip to content

[None][fix] fix logprob in trtllm-serve #6845

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
jhaotingc wants to merge 1 commit into
base: main
Choose a base branch
from
Open

[None][fix] fix logprob in trtllm-serve #6845

Show file tree
Hide file tree
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
2 changes: 1 addition & 1 deletion tensorrt_llm/serve/openai_protocol.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -589,7 +589,7 @@ def to_sampling_params(self, vocab_size: int = 32000) -> SamplingParams:

# TODO: migrate to use logprobs and prompt_logprobs
_return_log_probs=bool(self.logprobs),
)
logprobs=bool(self.logprobs))
return sampling_params

@model_validator(mode='before')
Expand Down
2 changes: 1 addition & 1 deletion tensorrt_llm/serve/postprocess_handlers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -68,7 +68,7 @@ def create_logprobs(token_ids: List[int],
# returning multiple logprobs is not supported
first_logprob = ChatCompletionLogProbsContent(
token=token,
logprob=max(logprob, -9999.0),
logprob=max(logprob[token_id].logprob, -9999.0),

@LinPoly LinPoly Aug 13, 2025

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

This will break TRT flow IIRC.

@jhaotingc jhaotingc Aug 13, 2025 •
edited
Loading

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

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

I didn't test, but that's the data format I got after I sent "logprobs": true to /v1/chat/completions, for both torch and trt sampler.

What's the way to sent the curl and have you checked if 6097 fix the issue (correctly get logprob from pytorch flow).

If you've verified the function works on 6097, we don't need this fix. Thanks!!

bytes=list(token.encode("utf-8", errors="replace")))
content.append(first_logprob)
chat_logprobs = ChatCompletionLogProbs(content=content)
Expand Down