July 11,2023 - Alibaba Cloud has unveiled its latest generative AI model for image generation for enterprise customers in China, underscoring growing efforts by hyperscalers to stake their claim on the nascent market for generative AI services.
Dubbed Tongyi Wanxiang, the model can respond to text prompts in Chinese and English to generate detailed images in an array of styles, encompassing watercolours, oil and Chinese painting to animation, sketch, flat illustration and 3D cartoons.
It can also transform any image into a new one with a similar style and stylise images through style transfer, which preserves the content of the original image while applying the visual style of another picture.
Powered by Alibaba Cloud’s capabilities in knowledge arrangement, visual AI and natural language processing, Tongyi Wanxiang leverages multilingual materials for enhanced training. It is touted to have robust semantic comprehension capability, resulting in more accurate and contextually relevant image generation.
Additionally, by optimising the high-resolution diffusion process based on the signal-to-noise ratio, the model can strike a balance between composition accuracy and detail sharpness while enhancing its ability to generate high-contrast, visually stunning images with clean backgrounds.
“Tongyi Wanxiang represents another significant milestone in our pursuit of advanced generative AI models as we continue to explore paradigm-shifting technologies that empower businesses and communities to unleash greater creativity and productivity,” said Zhou Jingren, chief technology officer of Alibaba Cloud.
“With the release of Tongyi Wanxiang, high-quality generative AI imagery will become more accessible, facilitating the development of innovative AI art and creative expressions for businesses across a wide range of sectors, including e-commerce, gaming, design and advertising,” Zhou added.
Separately, Alibaba Cloud has also introduced its ModelScopeGPT framework that lets enterprises and developers access and run a range of open-source large language models (LLMs) for performing sophisticated AI tasks, such as producing multilingual videos.
Alibaba Cloud’s latest generative AI services follows the earlier launch of its Tongyi Qianwen LLM that was developed based on Tongyi, a proprietary pre-trained model framework that unifies various AI models, including those that can turn text into images and short videos.
Tongyi Qianwen has since been integrated into Alibaba Cloud’s intelligent assistant, Tingwu, enabling the assistant to comprehend and analyse multimedia content with high levels of accuracy and efficiency. Over 360,000 users have accessed to the AI-powered assistant since its launch.
