Recently, Amazon Web Services launched a series of technology releases at the 2024 re: Invent Global Conference to support enterprise application of generative AI through full stack linkage innovation covering infrastructure, models, and applications, comprehensively reshaping the enterprise cloud innovation experience.
It is reported that Amazon Web Services has released three major areas of convergence: generative AI, data strategy, and cloud services. In the field of generative AI, it has launched the Amazon Nova series of basic models and strengthened multiple core services, accelerating enterprise application of generative AI innovation through lower training and inference costs, more model choices, and deeper application scenarios; In terms of data strategy, the release of the new generation Amazon SageMaker provides a unified platform for data, analytics, and AI; In terms of cloud services, we have launched a super server that provides real-time inference performance for trillion parameter models.
We not only continue to innovate at the core service level of the cloud, but also make breakthroughs in every technology stack from chips to models, and then to applications, enabling innovation at different levels to empower and evolve together. "Chen Xiaojian, General Manager of Amazon Web Services Greater China Product Department, said that large-scale innovation through full stack linkage can truly meet the development needs of today's enterprises and accelerate the value release of cutting-edge technologies.
Specifically, in the field of generative AI, Amazon Web Services comprehensively strengthens the three-layer technology stack of infrastructure, models, and applications, helping enterprises to more easily and economically apply generative AI to practical business scenarios. This update includes the launch of six basic models for Amazon Nova; Amazon Bedrock has newly integrated over 100 models and launched heavyweight updates such as AI protection, multi-agent collaboration, and model distillation, comprehensively optimizing the accuracy, cost, and response speed of inference scenarios; Amazon SageMaker AI will help businesses build, train, and deploy models faster and easier.
In the field of generative AI, we will see many companies move from the thinking stage to the practical stage in 2024, conducting a large number of scenario experiments. It is expected that by 2025, many companies will transition from the prototype verification stage to the production stage. At that time, enterprise needs will become more complex, not only in selecting models, but also in various technical support. Our goal in developing Amazon Bedrock is not only to provide a model market, but more importantly, to provide various productivity tools and production environment tools that enable models to reason and run at runtime. This is where the true value of Amazon Bedrock lies, "said Chen Xiaojian.
More and more companies are no longer using different data analysis tools in isolation, but are combining analytics, machine learning, and generative AI to gain insights. In response, Amazon Web Services has launched a new generation of Amazon SageMaker, which includes a new, unified studio that provides customers with a single data and AI development environment where users can find and access all data within their organization, and expand data and AI projects to different roles within the team for collaboration. In addition, Amazon Web Services continues to innovate in core areas such as computing, networking, storage, and databases, providing stronger underlying support for various workloads.