Spaces:

OFA-Sys

/

OFA-Generic_Interface

like 35

Build error

This is the demo of image captioning for OFA, which is a unified multimodal pretrained model that achieved SOTA performance in multiple downstream tasks. The paper is accepted by ICML 2022, and the github repo is popular in the community. See https://arxiv.org/abs/2202.03052 and https://github.com/OFA-Sys