Supermicro launches three NVIDIA-based generative AI SuperClusters for scalable LLM training.
Supermicro has launched three NVIDIA-based, full-stack, ready-to-deploy generative AI SuperClusters, which scale from enterprise to large LLM infrastructures. These Supermicro SuperCluster solutions are designed for powerful LLM training performance, large batch size, and high-volume workloads. The SuperClusters consist of air-cooled and liquid-cooled training and cloud-scale inference rack configurations, using the latest NVIDIA Tensor Core GPUs, networking, and NVIDIA AI Enterprise software.
March 18, 2024
17 Articles