亚马逊与Hugging Face合作，定制芯片助力AI模型低成本运行
近日，亚马逊公司的云部门宣布与人工智能初创公司Hugging Face达成合作。这项合作的目标是在亚马逊的定制计算芯片上更低成本地运行数千个AI模型。
Hugging Face，估值达45亿美元的初创公司，已成为AI研究人员和开发者分享聊天机器人或其他AI软件的中心。该公司得到了亚马逊、Alphabet旗下的谷歌和英伟达等公司的支持，是开发者获取和调整开源AI模型的主要平台，如Meta Platforms的Llama 3。
此次合作将使亚马逊和Hugging Face共同推动AI技术的普及和发展。亚马逊云部门将提供其定制的计算芯片，以支持Hugging Face的AI模型运行。这将大幅降低AI模型运行的成本，并为开发者提供更加高效的计算资源。
此次合作不仅有助于推动人工智能技术的进步，还将促进云计算和人工智能领域的深度融合。随着AI技术的不断发展，对于计算资源的需求也将不断增长。亚马逊和Hugging Face的合作将有助于满足这一需求，并推动AI技术的广泛应用。
英语如下：
News Title: Amazon Teams Up with Hugging Face to Power AI Models with Custom Chips for Lower Cost Operation
Keywords: Amazon, Hugging Face, AI Model Collaboration
News Content:
Amazon and Hugging Face Collaborate to Run AI Models on Custom Chips for Lower Cost Operation
Recently, Amazon Web Services (AWS) announced a collaboration with artificial intelligence startup Hugging Face. The aim of this collaboration is to run thousands of AI models on Amazon’s custom compute chips at lower costs.
Hugging Face, a startup valued at $4.5 billion, has become a hub for AI researchers and developers to share chatbots or other AI software. The company is supported by major companies such as Amazon, Google via Alphabet, and Nvidia, and is a leading platform for developers to access and adjust open-source AI models like Meta Platforms’ Llama 3.
This collaboration will push forward the popularization and development of AI technology by Amazon and Hugging Face. AWS will provide its custom compute chips to support the operation of Hugging Face’s AI models. This will significantly reduce the cost of running AI models and provide developers with more efficient computing resources.
This collaboration not only helps to drive the advancement of AI technology but also promotes the deep integration of cloud computing and artificial intelligence. As AI technology continues to develop, the demand for computing resources will also continue to grow. The collaboration between Amazon and Hugging Face will help meet this demand and promote the widespread application of AI.
【来源】https://www.ithome.com/0/770/049.htm
Views: 7