Tencent’s WeChat Vision team has introduced WeVisOne-VL, a new vision language model available in both 2B and 4B sizes. This innovative model consolidates various tasks including detection, optical character recognition (OCR), counting, segmentation, and alpha matting into a single framework.
WeVisOne-VL features a 9.6 million-parameter pixel decoder, enhancing its capability to process and analyze visual data efficiently. By unifying these functionalities, Tencent aims to streamline workflows for developers and researchers in the field of artificial intelligence.
This launch signifies Tencent’s commitment to advancing AI technologies, particularly in visual processing, and positions the company as a competitive player in the rapidly evolving landscape of AI models.
Compiled automatically by the Tech AI Newsdesk from public AI-news sources and summarised in our own words.
