On August 30, Ali Qwen released the second-generation visual language model Qwen2-VL, and the API of the flagship model Qwen2-VL-72B has been launched on the Ali Cloud Bailian platform. In a number of authoritative evaluations, some indicators of Qwen2-VL even surpassed closed-source models such as GPT-4o and Claude3.5-Sonnet. Compared to the previous generation model, Qwen2-VL can understand more than 20 minutes of long video, supporting video-based Q&A, dialogue and content creation applications; It can operate mobile phones and robots independently. With the ability of complex reasoning and decision-making, Qwen2-VL can be integrated into mobile phones, robots and other devices to automatically operate according to the visual environment and text instructions; Can understand multilingual text in image videos, including Chinese, English, most European languages, Japanese, Korean, Arabic, Vietnamese, etc.

来源：DIYuan