{"@context":"https://schema.org","@type":"Product","license":"https://creativecommons.org/licenses/by/4.0/","creditText":"Data: AintOnlyAI (https://aintonlyai.world), CC-BY-4.0","identifier":"nvidia-blackwell","name":"NVIDIA Blackwell","description":"NVIDIA Blackwell is a GPU architecture and platform designed for large-scale AI training and inference, succeeding the Hopper architecture (H100/H200). It operates at 1kW of power per GPU and requires direct liquid cooling, representing a new era for data center design. Blackwell powers AI factories for extreme-scale inference and is available across major cloud providers including AWS, Azure, and Oracle OCI.","url":"https://www.nvidia.com","sameAs":"https://www.nvidia.com","additionalType":"technology","areaServed":"United States","keywords":"semiconductors, compute, data_centers, ai_infrastructure_software, cooling, networking","additionalProperty":{"@type":"PropertyValue","name":"ai_adjacency_score","value":60}}