Skip to content
Open
Changes from 1 commit
Commits
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
1 change: 1 addition & 0 deletions scripts/start_gaudi_vllm_server.sh
Original file line number Diff line number Diff line change
Expand Up @@ -189,6 +189,7 @@ python3 -m vllm.entrypoints.openai.api_server \
--trust-remote-code \
--seed 2025 \
--distributed_executor_backend "${dist_backend}" \
--mm_processor_kwargs "max_pixels=1003520" \
Copy link

Choose a reason for hiding this comment

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

should this value be configurable by users and set 1003520 as default value?

Copy link
Author

Choose a reason for hiding this comment

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

If I change it to configurable, what option flag do you suggest I use?

Copy link

Choose a reason for hiding this comment

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

If I change it to configurable, what option flag do you suggest I use?

how about just reuse -e
cc @yangulei

Copy link
Author

Choose a reason for hiding this comment

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

I update the PR and try to reuse -e option to set --mm-processor-kwargs.
It is troublesome to process "max_pixles" alone.

Choose a reason for hiding this comment

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

If you want to pass --mm_processor_kwargs "max_pixels=1003520" to the vllm server, just pass -e "--mm_processor_kwargs max_pixels=1003520" to the start_gaudi_vllm_server.sh. We don't need to change the code.

Copy link
Author

Choose a reason for hiding this comment

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

We should better set a default value in script because user may doesn't know this options is needed.

Copy link
Author

Choose a reason for hiding this comment

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

I use -e opton, but with a default value for mm-processor-kwargs. How about this implementation?

"${extra_params[@]}")

echo "Start a vLLM server for ${model_name} on Gaudi $DEVICE_NAME with command:" |& tee -a "${log_file}"
Expand Down