--- title: "NVIDIA GB300 NVL72 | Aguasto" description: "Manufacturer reference for NVIDIA GB300 NVL72: Grace Blackwell Ultra, rack system, with sourced specifications, variant boundaries and deployment considerations." canonical: "https://www.aguasto.com/gpus/gb300-nvl72" last-updated: "2026-10-12" --- # NVIDIA GB300 NVL72 | Aguasto Canonical HTML: [NVIDIA GB300 NVL72](https://www.aguasto.com/gpus/gb300-nvl72) > Manufacturer reference for NVIDIA GB300 NVL72: Grace Blackwell Ultra, rack system, with sourced specifications, variant boundaries and deployment considerations. [Aguasto](https://www.aguasto.com/) [GPU Library](https://www.aguasto.com/gpus) NVIDIA GB300 NVL72 [Resources](https://www.aguasto.com/resources) [GPU Library](https://www.aguasto.com/gpus) [Model Index](https://www.aguasto.com/models) [Inference Stack](https://www.aguasto.com/inference-stack) [Learn](https://www.aguasto.com/learn) [Insights](https://www.aguasto.com/insights) [Research](https://www.aguasto.com/research) [Engineering tools](https://www.aguasto.com/tools) NVIDIA / Rack-scale system ## NVIDIA GB300 NVL72 NVIDIA reports GB300 NVL72 shipped in 2025; a specific OEM configuration and delivery date still require confirmation. Original Aguasto conceptual illustration. Not manufacturer geometry, an operating Aguasto asset, or a measured benchmark. Aguasto Research documents third-party hardware, models and software. References do not represent Aguasto inventory, evaluated deployments or available inference services. ### Manufacturer specification reference Architecture Grace Blackwell Ultra[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) · [NVIDIA Blackwell Ultra datasheet (external source)](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) · [NVIDIA Vera Rubin POD: Seven Chips, Five Rack-Scale Systems, One AI Supercomputer (external source)](https://developer.nvidia.com/blog/nvidia-vera-rubin-pod-seven-chips-five-rack-scale-systems-one-ai-supercomputer/) · [CUDA GPU Compute Capability (external source)](https://developer.nvidia.com/cuda/gpus) Category Complete rack-scale reference[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) · [NVIDIA Blackwell Ultra datasheet (external source)](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) · [NVIDIA Vera Rubin POD: Seven Chips, Five Rack-Scale Systems, One AI Supercomputer (external source)](https://developer.nvidia.com/blog/nvidia-vera-rubin-pod-seven-chips-five-rack-scale-systems-one-ai-supercomputer/) · [CUDA GPU Compute Capability (external source)](https://developer.nvidia.com/cuda/gpus) Accelerator memory 20,000 GBPublished rounded 20 TB aggregate GPU memory converted to decimal GB; excludes CPU LPDDR5X.[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) Memory type HBM3e[NVIDIA Blackwell Ultra datasheet (external source)](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) Memory bandwidth 576 TB/sUp to rack aggregate GPU memory bandwidth, not NVLink bandwidth.[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) Power Not verified for this recordTotal rack operating draw not verified in the reviewed sources; individual GPU TDP is not rack power.[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) · [NVIDIA Blackwell Ultra datasheet (external source)](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) · [NVIDIA Vera Rubin POD: Seven Chips, Five Rack-Scale Systems, One AI Supercomputer (external source)](https://developer.nvidia.com/blog/nvidia-vera-rubin-pod-seven-chips-five-rack-scale-systems-one-ai-supercomputer/) · [CUDA GPU Compute Capability (external source)](https://developer.nvidia.com/cuda/gpus) Precision formats FP4, FP8, FP6, INT8, FP16, BF16, TF32, FP32, FP64[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) Cooling / integration Fully liquid-cooled rack-scale architecture.[NVIDIA GB300 NVL72 (external source)](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) · [NVIDIA Blackwell Ultra datasheet (external source)](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) · [NVIDIA Vera Rubin POD: Seven Chips, Five Rack-Scale Systems, One AI Supercomputer (external source)](https://developer.nvidia.com/blog/nvidia-vera-rubin-pod-seven-chips-five-rack-scale-systems-one-ai-supercomputer/) · [CUDA GPU Compute Capability (external source)](https://developer.nvidia.com/cuda/gpus) ### Configuration boundaries - GPU memory and CPU memory are separate. Fast memory on the product page includes both. - Rack totals are manufacturer-published rounded figures; no inferred per-GPU memory value is supplied. ### Published arithmetic throughput Manufacturer peaks, not Aguasto measurements or tokens per second **Specification** FP4 peak compute **Precision** FP4 **Published peak** 1,080 PFLOPS **Boundary & source** Whole rack, manufacturer peak dense Tensor Core specification[Manufacturer source ↗](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) **Specification** FP4 peak compute **Precision** FP4 **Published peak** 1,440 PFLOPS **Boundary & source** Whole rack, manufacturer peak sparse Tensor Core specification[Manufacturer source ↗](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) ### Deployment considerations - Verify CPU/GPU software architecture assumptions and end-to-end serving topology. - Use measured workload power and OEM site-planning documentation before calculating facility capacity. #### Software ecosystem CUDA · NVIDIA Mission Control · NVIDIA AI Enterprise. Confirm the exact model, runtime release, server integration and vendor support matrix before evaluating a deployment. ### Continue the research [NVIDIA Dynamo](https://www.aguasto.com/inference-stack/dynamo) [TensorRT-LLM](https://www.aguasto.com/inference-stack/tensorrt-llm) [Kubernetes GPU orchestration — NVIDIA device plugin](https://www.aguasto.com/inference-stack/kubernetes) [Model weights versus runtime memory](https://www.aguasto.com/learn/model-weights-and-runtime-memory) [Compare compatible hardware categories](https://www.aguasto.com/gpus/compare) [Why modular infrastructure?](https://www.aguasto.com/why-modular) ### Sources & verification Reference revision 1 · Reviewed October 12, 2026. Manufacturer claims and documentation are attributed; Aguasto has not run the cited products in production. - [NVIDIA GB300 NVL72 ↗](https://www.nvidia.com/en-us/data-center/gb300-nvl72/) 72 Blackwell Ultra GPUs and 36 Grace CPUs; 20 TB GPU memory; Up to 576 TB/s aggregate GPU bandwidth; Precision table; Mission Control. Publication: date not stated in the reviewed source. Verified: 2026-10-12. - [NVIDIA Blackwell Ultra datasheet ↗](https://dam-cdn.nvd.orangelogic.com/AssetLink/1k0p832eq8r5ca0u5383ie5o4tp3bst1.pdf) Fully liquid-cooled GB300 rack; HBM3e memory type; AI Enterprise software ecosystem. Publication: date not stated in the reviewed source. Verified: 2026-10-12. - [NVIDIA Vera Rubin POD: Seven Chips, Five Rack-Scale Systems, One AI Supercomputer ↗](https://developer.nvidia.com/blog/nvidia-vera-rubin-pod-seven-chips-five-rack-scale-systems-one-ai-supercomputer/) GB200 NVL72 first shipments in 2024; GB300 NVL72 shipments in 2025. Publication: date not stated in the reviewed source. Verified: 2026-10-12. - [CUDA GPU Compute Capability ↗](https://developer.nvidia.com/cuda/gpus) CUDA product identification for H100, H200, B200, B300, GB200 and GB300. Publication: date not stated in the reviewed source. Verified: 2026-10-12. ### Reference journey [Compare hardware ↗](https://www.aguasto.com/gpus/compare) [Model Index ↗](https://www.aguasto.com/models) [Serving software ↗](https://www.aguasto.com/inference-stack) [OEM collaboration ↗](https://www.aguasto.com/partners) [Sources & verification ↓](https://www.aguasto.com/gpus/gb300-nvl72#sources) Research reference · No available Aguasto service ## Company navigation - [Home](https://www.aguasto.com/) - [Infrastructure](https://www.aguasto.com/infrastructure) - [GPU Cloud](https://www.aguasto.com/gpu-cloud) - [AI Inference](https://www.aguasto.com/ai-inference) - [Development & Investment](https://www.aguasto.com/development) - [Company](https://www.aguasto.com/company) - [Contact](https://www.aguasto.com/contact) - [Privacy](https://www.aguasto.com/privacy) - [Resources](https://www.aguasto.com/resources) - [GPU Library](https://www.aguasto.com/gpus) - [Model Index](https://www.aguasto.com/models) - [Inference Stack](https://www.aguasto.com/inference-stack) - [Learn](https://www.aguasto.com/learn) - [Insights](https://www.aguasto.com/insights) - [Research](https://www.aguasto.com/research) - [Industry cases](https://www.aguasto.com/research/cases) - [Why Modular](https://www.aguasto.com/why-modular) - [Partners](https://www.aguasto.com/partners) - [GPU comparison](https://www.aguasto.com/gpus/compare) - [Model comparison](https://www.aguasto.com/models/compare) - [Engineering tools](https://www.aguasto.com/tools) - [Development friction](https://www.aguasto.com/research/hyperscale-development-friction) - [NVIDIA H100 SXM](https://www.aguasto.com/gpus/h100) - [NVIDIA H200 SXM](https://www.aguasto.com/gpus/h200) - [NVIDIA B200 SXM](https://www.aguasto.com/gpus/b200) - [NVIDIA B300](https://www.aguasto.com/gpus/b300) - [NVIDIA GB200 NVL72](https://www.aguasto.com/gpus/gb200-nvl72) - [NVIDIA GB300 NVL72](https://www.aguasto.com/gpus/gb300-nvl72) - [NVIDIA Vera Rubin NVL72](https://www.aguasto.com/gpus/vera-rubin-nvl72) - [AMD Instinct MI300X](https://www.aguasto.com/gpus/mi300x) - [AMD Instinct MI325X](https://www.aguasto.com/gpus/mi325x) - [AMD Instinct MI355X](https://www.aguasto.com/gpus/mi355x) - [AMD Instinct MI455X](https://www.aguasto.com/gpus/mi455x) - [AMD Helios](https://www.aguasto.com/gpus/helios) - [GLM-4.5](https://www.aguasto.com/models/glm-4-5) - [Qwen3-235B-A22B](https://www.aguasto.com/models/qwen3-235b-a22b) - [DeepSeek-R1](https://www.aguasto.com/models/deepseek-r1) - [Llama 3.3 70B Instruct](https://www.aguasto.com/models/llama-3-3-70b-instruct) - [Gemma 3 27B IT](https://www.aguasto.com/models/gemma-3-27b-it) - [gpt-oss-120b](https://www.aguasto.com/models/gpt-oss-120b) - [Kimi-K2-Instruct](https://www.aguasto.com/models/kimi-k2-instruct) - [vLLM](https://www.aguasto.com/inference-stack/vllm) - [SGLang](https://www.aguasto.com/inference-stack/sglang) - [TensorRT-LLM](https://www.aguasto.com/inference-stack/tensorrt-llm) - [NVIDIA Dynamo](https://www.aguasto.com/inference-stack/dynamo) - [NVIDIA Triton Inference Server](https://www.aguasto.com/inference-stack/triton) - [AMD ROCm](https://www.aguasto.com/inference-stack/rocm) - [Ray Serve](https://www.aguasto.com/inference-stack/ray-serve) - [Kubernetes GPU orchestration — NVIDIA device plugin](https://www.aguasto.com/inference-stack/kubernetes) - [Model weights are only the beginning of the memory budget](https://www.aguasto.com/learn/model-weights-and-runtime-memory) - [Why tokens per second is not enough to benchmark inference](https://www.aguasto.com/research/inference-benchmark-methodology) - [Prefill, decode and the cost of separating them](https://www.aguasto.com/learn/prefill-decode-and-disaggregated-serving) - [Choosing between vLLM, SGLang and TensorRT-LLM](https://www.aguasto.com/learn/choosing-an-inference-serving-framework) - [From Power to Inference—Building a Distributed AI Compute Network](https://www.aguasto.com/learn/from-power-to-inference) - [Dedicated compute and shared inference: compare the whole workload](https://www.aguasto.com/insights/dedicated-and-shared-inference-economics) - [Distributed inference reliability begins with the failure boundary](https://www.aguasto.com/learn/distributed-inference-reliability) - [Core Scientific: reading contracted, energized and billed capacity](https://www.aguasto.com/research/cases/core-scientific-capacity-disclosures) - [Why Modular AI Infrastructure Matters Now](https://www.aguasto.com/research/why-modular-ai-infrastructure-matters-now) - [Official LinkedIn](https://www.linkedin.com/company/aguasto) This is a generated representation of the public HTML page. Diagrams are conceptual. GPU and inference services are planned and are not currently available to book or call. The contact page describes the available inquiry flow.