Automatically creating description sentences for images is a task that involves aligning image under-standing with natural language processing. This paper presents a model for image captioning that employs a vision Transformer model with re-attention as the encoder and a T5-based model as the decoder. We conduct exper-iments with several Transformer-based models for image captioning on the Flickr8k datasets in both English and Vietnamese. The DeepViT+ViT5 model achieves a BLEU-4 score of 37.98, which is the best result so far for Vietnamese image captioning on the Flickr8k dataset.