New AI Models Released by Alibaba to Enhance Vision Language Capabilities

Alibaba, one of China’s leading tech companies, has recently unveiled two groundbreaking AI models that have the potential to revolutionize the field of artificial intelligence. Known as Qwen-VL and Qwen-VL-Chat, these open-source models are primarily vision language models, allowing them to “read” and understand images rather than text. This sets them apart from other competing models like Chat-GPT and Google Bard.

Qwen-VL-Chat, in particular, boasts a wide range of complex features. It can analyze street signs to provide directions, solve math problems by examining images, and even construct cohesive narratives based on multiple pictures. For example, it can translate Mandarin text on a hospital sign into English or assist news organizations in generating accurate photo captions.

Meanwhile, Qwen-VL represents an upgraded version of Alibaba’s existing image-reading chatbot, now capable of processing higher-resolution images. This advancement opens up new possibilities for assisting visually impaired individuals, enabling them to scan items and have the chatbot describe the contents to them.

As part of Alibaba’s commitment to collaboration and innovation, both models will be available on Alibaba Cloud’s Modelscope platform and Hugging Face, a popular startup with an extensive collection of AI models.

Alibaba’s introduction of these new AI models is a significant step in the ongoing race among developers to create increasingly sophisticated tools. The technology is no longer seen as mere novelty but a genuine game-changer with wide-ranging applications. By releasing the models as open-source, Alibaba empowers users to customize and employ these tools for their specific purposes, without the need to build large language models from scratch.

The Chinese government recognizes the critical importance of AI development and has recently issued comprehensive regulations for the field. These regulations have given companies like Alibaba the green light to bring their AI products to the public. Alibaba’s dedication to AI is evident in its plans to restructure and spin off its cloud computing division, Alibaba Cloud, into an independent entity. The inclusion of AI research within the cloud division will enhance efficiency and drive further advancements in the technology.

While Chinese tech companies are slightly behind their American counterparts in terms of model size, the Chinese government views AI as a key component of its technological future. This has led to an intense competition between China and the U.S. to dominate the field. Both countries recognize the potential military and surveillance implications of AI technology and strive to maintain a leading position in its development.


Q: What are the names of Alibaba’s new AI models?
A: Alibaba’s latest AI models are called Qwen-VL and Qwen-VL-Chat.

Q: What distinguishes Qwen-VL-Chat from other models?
A: Qwen-VL-Chat is a vision language model that can read images and perform various complex tasks like translating text or generating photo captions.

Q: How can Qwen-VL assist visually impaired individuals?
A: Qwen-VL can analyze images at higher resolutions, allowing it to describe the contents of scanned items to visually impaired users.

Q: Where can these new AI models be accessed?
A: The models will be available on Alibaba Cloud’s Modelscope platform and the startup Hugging Face.

Q: Why is AI development considered a priority by the Chinese government?
A: The Chinese government recognizes the potential of AI technology for the nation’s technological future and has issued comprehensive regulations to support its development.