One-click deployment of NVIDIA\u2019s open-source inference framework across public, private, hybrid, and on-prem environments<\/p>\n
Gcore, the global infrastructure and software provider for AI, cloud, network, and security solutions, today announced the integration of\u00a0NVIDIA Dynamo\u00a0into its AI inference solutions. The integration delivers significant GPU efficiency gains\u2014up to 6x higher throughput and 2x lower latency\u2014as a fully managed, one-click deployment. Dynamo is available now on Gcore Everywhere Inference and Gcore Everywhere AI.<\/p>\n
NVIDIA Dynamo is an open-source inference framework, specifically designed to accelerate and optimize large-scale generative AI and inference models. Dynamo addresses the core challenges that businesses experience when running inference at scale: GPU underutilization, static resource allocation, memory bottlenecks, and data transfer inefficiency.<\/p>\n
Gcore is delivering Dynamo as a fully managed solution, pre-optimized for popular inference models. Customers can activate Dynamo with a single click within the Gcore Customer Portal, without managing routing, KV cache logic, or GPU scheduling. This builds on Gcore\u2019s commitment to simplifying AI deployment through its intuitive, easy-to-use platform. The Dynamo integration is supported across private cloud, hybrid, and on-premises inference environments on Gcore Everywhere AI and Everywhere Inference.<\/p>\n
Seva Vayner, Product Director of Edge Cloud and AI at Gcore,\u00a0comments: \u201cModern inference isn\u2019t just \u2018run a model\u2019\u2014it\u2019s batching, routing, dynamic workloads, longer contexts, and tight SLOs. In that reality, small scheduling and utilization losses become big performance and cost penalties. By integrating Dynamo as a managed service in Gcore, we bring advanced GPU optimization directly into the runtime path so customers see higher effective throughput and steadier tail latency, without operating the complexity themselves.\u201d<\/p>\n
Beyond performance gains, NVIDIA Dynamo delivers meaningful cost optimization by increasing GPU utilization and reducing wasted cycles during decode and cache recomputation. By disaggregating prefill and decode, applying KV cache-aware routing, and leveraging NIXL for efficient inter-node communication, Dynamo ensures more requests are processed on the same hardware. This lowers cost per\u00a0token\u00a0and improves overall ROI. Gcore makes it particularly easy to access these efficiencies at scale.<\/p>\n
Dynamo-powered inference is available today on Gcore Inference and Everywhere AI. Visit Gcore at MWC (Barcelona, March 2\u20135) or at GTC (San Jose, March 16\u201319) for an in-person demonstration of NVIDIA Dynamo on Gcore.The post Gcore Integrates NVIDIA Dynamo for AI Inference first appeared on AI-Tech Park.<\/p>\n","protected":false},"excerpt":{"rendered":"
One-click deployment of NVIDIA\u2019s open-source inference framework across public, private, hybrid, and on-prem environments Gcore, the global infrastructure and software provider for AI, cloud, network, and security solutions, today announced the integration of\u00a0NVIDIA Dynamo\u00a0into its AI inference solutions. The integration delivers significant GPU efficiency gains\u2014up to 6x higher throughput and…<\/p>\n
The post Gcore Integrates NVIDIA Dynamo for AI Inference<\/a> first appeared on AI-Tech Park<\/a>.<\/p>\n<\/div>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"site-container-style":"default","site-container-layout":"default","site-sidebar-layout":"default","disable-article-header":"default","disable-site-header":"default","disable-site-footer":"default","disable-content-area-spacing":"default","footnotes":""},"categories":[4,1],"tags":[3],"class_list":["post-1609","post","type-post","status-publish","format-standard","hentry","category-ai","category-ai-and-ml","tag-ai"],"yoast_head":"\n