# Multi-Scenario Applicability

This is a lightweight image captioning model based on bert-tiny and vit-tiny, weighing only 40MB, with extremely fast inference speed on CPU.

ViTPose++ is a vision Transformer-based foundational model for human pose estimation, achieving an outstanding performance of 81.1 AP on the MS COCO keypoint test set.

A vision-language model based on VIT image encoder and distilled GPT-2 text decoder for image caption generation tasks

A spell checker built on the T5-Base transformer for detecting and correcting text spelling errors.