{"id":1651,"date":"2026-06-22T14:49:39","date_gmt":"2026-06-22T14:49:39","guid":{"rendered":"https:\/\/rdp.in\/gpu-mart\/product\/nvidia-gb300-nvl72-rack-scale-ai-factory\/"},"modified":"2026-06-24T14:40:40","modified_gmt":"2026-06-24T14:40:40","slug":"nvidia-gb300-nvl72-rack-scale-ai-factory","status":"publish","type":"product","link":"https:\/\/rdp.in\/gpu-mart\/product\/nvidia-gb300-nvl72-rack-scale-ai-factory\/","title":{"rendered":"NVIDIA GB300 NVL72 Rack-Scale AI Factory"},"content":{"rendered":"<p>The NVIDIA GB300 NVL72 Rack-Scale AI Factory is a liquid-cooled, rack-scale AI infrastructure system built on the NVIDIA Blackwell Ultra platform. It unifies 72 Blackwell Ultra GPUs and 36 Grace CPUs into a single NVLink domain, so an entire rack operates as one accelerator for trillion-parameter training and test-time-scaling inference. Delivered, integrated and supported across India by RDP Technologies Limited \u2014 predictable INR pricing, GST invoice and pan-India onsite support.<\/p>\n<h3>Key highlights<\/h3>\n<ul>\n<li>72 NVIDIA Blackwell Ultra GPUs and 36 NVIDIA Grace CPUs unified as a single NVLink domain \u2014 the rack behaves as one accelerator.<\/li>\n<li>130 TB\/s NVLink scale-up bandwidth across all 72 GPUs.<\/li>\n<li>20 TB GPU memory, 37 TB fast memory and 17 TB LPDDR5X CPU memory for trillion-parameter models and long-context inference.<\/li>\n<li>2,592 Arm Neoverse V2 cores (36 Grace CPUs) for data orchestration and CPU-side pipelines.<\/li>\n<li>Fully liquid-cooled rack-scale architecture for sustained high-density performance and energy efficiency.<\/li>\n<li>Engineered for AI factories: training, fine-tuning, high-throughput inference, test-time scaling and agentic \/ reasoning systems.<\/li>\n<li>Delivered as a validated reference-architecture rack \u2014 onsite integration, deployment and pan-India support by RDP.<\/li>\n<li>Make-in-India delivery; GST invoice; available on GeM for government and PSU procurement.<\/li>\n<\/ul>\n<h3>AI workload positioning<\/h3>\n<p>The GB300 NVL72 is built to be an AI factory, not a single server. With 20 TB of GPU memory, a 130 TB\/s NVLink scale-up fabric and 54 TB of total fast and CPU memory across 72 Blackwell Ultra GPUs and 36 Grace CPUs, one rack can train and serve trillion-parameter models, run test-time-scaling (reasoning) inference and host large agentic systems that would otherwise require a room of disaggregated servers. Because the whole rack is one NVLink domain, parallelism scales up rather than out, removing network bottlenecks at frontier scale.<\/p>\n<h3>Built for AI factories \u2014 who it is for<\/h3>\n<ul>\n<li>Enterprise AI labs and frontier-model teams<\/li>\n<li>Cloud, neocloud and hosting providers building AI capacity<\/li>\n<li>Government and sovereign AI programmes<\/li>\n<li>Research organisations and national HPC \/ AI centres<\/li>\n<li>Large IT and infrastructure teams standing up AI factories<\/li>\n<li>System integrators and hyperscale AI operations<\/li>\n<\/ul>\n<h3>Why it matters<\/h3>\n<ul>\n<li><strong>Rack-scale integration<\/strong> \u2014 delivered as a validated reference-architecture rack, not a parts list.<\/li>\n<li><strong>Scale-up fabric<\/strong> \u2014 a single NVLink domain so parallelism scales up, not out, removing network bottlenecks.<\/li>\n<li><strong>Very high memory capacity<\/strong> \u2014 20 TB GPU + 37 TB fast + 17 TB LPDDR5X for the largest models and longest contexts.<\/li>\n<li><strong>Liquid cooling<\/strong> \u2014 sustained high-density performance with better energy efficiency and thermal headroom.<\/li>\n<li><strong>Enterprise deployment readiness<\/strong> \u2014 onsite integration, support and financing options for production AI factories.<\/li>\n<\/ul>\n<h3>Buy with confidence<\/h3>\n<ul>\n<li><strong>Make-in-India OEM<\/strong> \u2014 designed, integrated and supported in India; 14 years and 100,000+ devices shipped.<\/li>\n<li><strong>Pan-India onsite support<\/strong> with reference-architecture integration and deployment.<\/li>\n<li><strong>Transparent INR pricing<\/strong> \u2014 GST invoice, eligible for input-tax credit; financing and as-a-service options available.<\/li>\n<li><strong>Sovereign-ready<\/strong> \u2014 India data residency (DPDP), MeitY-recognised, ISO 27001 \/ SOC 2 deployment paths.<\/li>\n<li><strong>Government procurement<\/strong> \u2014 available on GeM for PSU and government AI programmes.<\/li>\n<li><strong>Talk to a solution architect<\/strong> \u2014 a no-cost reference-architecture consultation to design and size your AI factory.<\/li>\n<\/ul>\n<p><strong>Price on application.<\/strong> Configured and quoted per deployment (power, cooling, fabric and financing). Talk to a solution architect to design your AI factory.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>72 Blackwell Ultra GPUs \u00b7 36 Grace CPUs \u00b7 20 TB GPU memory \u00b7 liquid-cooled NVL72 rack<\/p>\n","protected":false},"featured_media":1660,"comment_status":"open","ping_status":"closed","template":"","meta":[],"product_brand":[],"product_cat":[17],"product_tag":[181,185,187,183,182,184,186,180],"class_list":["post-1651","product","type-product","status-publish","has-post-thumbnail","product_cat-rack-scale-ai-systems","product_tag-ai-factory","product_tag-blackwell-ultra","product_tag-contact-for-quote","product_tag-gb300","product_tag-liquid-cooled","product_tag-nvl72","product_tag-nvlink","product_tag-rack-scale","pa_budget-band-1cr","pa_form-factor-rack","pa_gpu-model-blackwell-ultra","pa_industry-media-gaming-entertainment","pa_industry-neocloud","pa_industry-public-sector-sovereign-ai","pa_industry-research-higher-education","pa_use-case-generative-ai","pa_use-case-hpc-ai","pa_use-case-inference","pa_use-case-llm-training","pa_use-case-fine-tuning","pa_use-case-nlp-speech","pa_use-case-sovereign-ai","pa_workload-fit-training-inference","first","onbackorder","taxable","shipping-taxable","purchasable","product-type-simple"],"_links":{"self":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product\/1651","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product"}],"about":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/types\/product"}],"replies":[{"embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/comments?post=1651"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/media\/1660"}],"wp:attachment":[{"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/media?parent=1651"}],"wp:term":[{"taxonomy":"product_brand","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_brand?post=1651"},{"taxonomy":"product_cat","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_cat?post=1651"},{"taxonomy":"product_tag","embeddable":true,"href":"https:\/\/rdp.in\/gpu-mart\/wp-json\/wp\/v2\/product_tag?post=1651"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}