benchmark icon indicating copy to clipboard operation
benchmark copied to clipboard

Published 20 hours ago • verified publisher icon pytorch

Llava model is not running its vision module

Open BowenBao opened this issue 11 months ago • 1 comments

In order for that to happen, an additional input pixel_values must be provided. This is different from standard HuggingFaceModel.

BowenBao avatar Feb 28 '24 01:02 BowenBao