Add image-text-to-text pipeline tag
#2
by
nielsr
HF staff
- opened
This PR updates the model card metadata to use the image-text-to-text pipeline tag. This tag better reflects the model's multimodal capabilities, including image captioning and visual question answering, as demonstrated in the provided examples and described in the paper. This change improves the model's discoverability on the Hub for users seeking vision-language models.