Meta公司近日宣布，将启用两个全新的24,576个GPU的超级集群，用于训练其最新的大语言模型Llama3。这两个基于AI Research SuperCluster (RSC)的集群，是Meta在2022年推出的，它们将在自然语言处理、语音识别和图像生成等AI研究领域发挥重要作用。与之前的包含16,000个Nvidia A100 GPU的集群相比，新集群的性能有了显著提升，为Meta在生成式AI产品开发方面提供了更强大的支持。
英文标题：Meta Reveals Details of Two New 24k GPU Clusters for Training Llama3
英文关键词：Meta, AI Research SuperCluster (RSC), Llama 3, GPU, AI Infrastructure, Natural Language Processing, Speech Recognition, Image Generation
英文新闻内容：
Meta, the social media giant, has announced the deployment of two new 24,576 GPU superclusters to train its latest large language model, Llama3. These clusters, based on Meta’s AI Research SuperCluster (RSC) launched in 2022, will play a pivotal role in advancing research in areas such as natural language processing, speech recognition, and image generation. Compared to the previous cluster with 16,000 Nvidia A100 GPUs, the new clusters offer a significant performance boost, laying the groundwork for advancements in Meta’s generative AI product development.
【来源】https://www.datacenterdynamics.com/en/news/meta-reveals-details-of-two-new-24k-gpu-ai-clusters/
Views: 1