Alibaba Cloud unveiled a series of AI products and services on Tuesday to meet corporate customers’ burgeoning demand for generative AI development.
The product suite, spanning AI computing platforms to AI acceleration services, is designed to help customers build their own generative AI applications more efficiently and cost-effectively, said the cloud computing company at its annual 2023 Alibaba Cloud Global Summit.
The launches coincide with a surging global generative AI market. According to a Bloomberg Intelligence report, generative AI is poised to expand its impact from less than 1% of total IT hardware, software services, advertising, and gaming market spending to 10% by 2032.
“Our comprehensive AI services are designed to help customers seize the unparalleled opportunities offered by this unstoppable technology trend,” said Selina Yuan, President of International Business at Alibaba Cloud, in a release.
At the global summit, Alibaba Cloud said its AI computing platform, PAI-Lingjun Intelligence Computing Service, will be available in Singapore early next year, followed by more markets in Asia throughout the year.
The platform, already in use in Mainland China, offers corporate customers a wide range of AI capabilities, including large-scale deep learning and intelligent computing, AI development and computing resource management.
Large-scale training for generative AI models can be time-consuming. On top of that, inference, which refers to the process of running the AI models, is computationally intensive as well. To accelerate the training and inference of AI models, Alibaba Cloud has launched an AI acceleration solution that optimizes the utilization of computing instances to expedite AI model training and inference tasks.
The solution can speed up AI training models by up to 70% and inference by up to three times, according to a deep learning benchmark, Stanford DAWN Deep Learning Benchmark, widely recognized in the industry.
Besides the underlying infrastructure and platform, Alibaba Cloud also made a foray into generative AI applications with the launch of a large language model-based conversational search engine.
The engine allows enterprises to build a dedicated conversational search system and supports multimodal search results such as answers, URLs, and images to cater to business search requirements in e-commerce, multimedia, social media, and big data queries.
“As part of our commitment to foster more inclusive and innovative application development for generative AI, we have been joining forces with industry leaders to further extend our capabilities for different scenarios,” said Yuan.
The cloud computing company said it has collaborated with software provider Red Hat to make Red Hat OpenShift, a platform for building, deploying and managing applications at scale accessible on Alibaba Cloud.