-
-
Notifications
You must be signed in to change notification settings - Fork 4.8k
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.
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
[Bugfix] Update InternVL input mapper to support image embeds #9351
[Bugfix] Update InternVL input mapper to support image embeds #9351
Conversation
👋 Hi! Thank you for contributing to the vLLM project. Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging. To run CI, PR reviewers can do one of these:
🚀 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks for the fix!
…roject#9351) Signed-off-by: charlifu <[email protected]>
…roject#9351) Signed-off-by: Vinay Damodaran <[email protected]>
…roject#9351) Signed-off-by: Alvant <[email protected]>
…roject#9351) Signed-off-by: Amit Garg <[email protected]>
…roject#9351) Signed-off-by: Sumit Dubey <[email protected]>
…roject#9351) Signed-off-by: Maxime Fournioux <[email protected]>
The InternVL input mapper originally only outputs MultiModalInputs with pixel values. However, the model itself has support for image embedding inputs and this change is needed for image embedding inputs to be recognized by
_parse_and_validate_image_input
(and thus the forward function).FIX bug found locally.