Multimodal prompting

Hi,

with the URL, I get response like
The image depicts three llamas, one of which is purple and wearing a party hat, standing behind a table with a glass of beer on it.

but with the local image,
The image depicts three llamas, one of which is wearing a party hat, sitting at a table with a glass of liquid in front of them.

why with the same image content in both url and local image, on gives beer and another says liquid, these response are consistent, url always returns beer where as local image returns liquid
why?

hi @Ram_Kumar_Koppu

can you share both image the url and the local image you got?

it is probably the url image is using different version of llama than the local image provided in the metadata files(File==>Open==>image==>Repo.png

Regards
DP