On October 15, 2024, NVIDIA announced it was contributing selected electro-mechanical design elements from its GB200 NVL72 rack-scale system to the Open Compute Project (OCP). The contribution covers rack and tray design, power and liquid-cooling details, and NVLink cable-cartridge dimensions—not the entire GB200 system or all of its software.
What NVIDIA contributed
NVIDIA’s announcement at the 2024 OCP Global Summit described sharing “key portions” of the GB200 NVL72 system’s electro-mechanical design. OCP later detailed the contribution as building on its Open Rack V3 (ORv3) specifications, with design elements for rack architecture, bus bars, cabling, cooling, and compute and switch trays.
OCP’s description includes these engineering details:
- Rack architecture: a reinforced design supporting 19-inch EIA equipment.
- Power distribution: a high-capacity bus bar rated at 1400A in OCP’s description. This is a design specification, not a measured performance result.
- NVLink cabling: cable-cartridge design details, including physical volumetrics.
- Liquid cooling: blind-mate connections and manifolds.
- Compute and switch trays: liquid-cooled MGX trays in a 1RU form factor.
These are features described in OCP’s account of the contribution; they should not be mistaken for independent test results or for proof that each feature is a separately approved standard.
#1 Best Overall
- PLEASE NOTE: Exporting an NVIDIA RTX Pro 6000 GPU outside the US requires strict adherence to the U.S. Export Administration Regulations (EAR) and issuance of an export license from the Bureau of Industry and Security (BIS). Compliance and Know Your Customer (KYC) screening may be required as a condition of order acceptance. [NVIDIA Blackwell Streaming Multiprocessor] The new SM features increased processing throughput, and new neural shaders that integrate neural networks inside of programmable shaders | DLSS 4: Multi Frame Generation ensures ultra-smooth frame pacing for lifelike simulations.
- [Double-Flow-Through Design] The RTX PRO 6000 Blackwell features a double-flow-through cooling design, optimizing efficiency and airflow to sustain peak performance under 600W power loads. | [5th Gen Tensor Cores] Deliver up to 3X the performance of the previous generation and support for FP4 precision for faster AI model processing times with reduced memory usage, enabling local fine-tuning of LLMs and generative AI | [4th Gen Ray Tracing Cores] Double the ray-triangle intersection rate of the previous generation to create photoreal, physically accurate scenes and immersive 3D designs with RTX Mega Geometry, which enables up to 100X more ray-traced triangles.
- [PCIe Gen 5] Support for PCIe Gen 5 provides double the bandwidth of PCIe Gen 4, improving data-transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks like AI, data science, and 3D modeling. | [GDDR7 Memory] With 96 GB of GPU memory and 1.8 TB ps bandwidth, it can tackle massive 3D and AI projects, fine-tune AI models locally, explore large-scale VR environments, and drive larger multi-app workflows.
- [DisplayPort 2.1] Achieve unparalleled visual clarity and performance, driving high resolution displays at up to 8K at 240 Hz and 16K at 60 Hz. Increased bandwidth enables seamless multi-monitor setups while HDR and higher color depth support ensures superior color accuracy for precision work, such as video editing, 3D design, and live broadcasting.
- [Universal MIG] Divide a single RTX PRO 6000 Blackwell into multiple isolated instances, each with dedicated resources, allowing for concurrent execution of multiple workloads, optimized GPU utilization, and secure isolation of different applications or users. [WARRANTY] 3 YR Manufacturer's Warranty. Bulk OEM Packaging. Retail Packaging is NOT included.
What the contribution means—and what it does not
The contribution gives infrastructure suppliers design information they can use when developing systems around NVIDIA’s rack-scale Blackwell platform. NVIDIA named system partners working on Blackwell servers, indicating ecosystem participation. That announcement alone does not confirm that every partner design has shipped, is available in a particular configuration, or has passed independent qualification.
Nor does the announcement say NVIDIA open-sourced the entire GB200 NVL72 system, its software, or every implementation detail. The supported description is narrower: selected foundational electro-mechanical elements were contributed to OCP.
Rank #2
- Professional GPU with Blackwell Architecture
- Blackwell Architecture
- 24GB GDDR7 with PCIe 5.0 & Ray Tracing
- AI Workstation
How the OCP specification record fits in
OCP lists an “MGX Accelerated Computing Rack and Trays” specification authored by NVIDIA with an effective date of August 29, 2024. That catalog entry is a document record, not a substitute for checking the current revision. Anyone designing or qualifying a deployment should consult the current OCP specification and confirm which revision applies.
Related system and facility work
NVIDIA also described a joint GB200 NVL72 reference architecture with Vertiv focused on data-center power and cooling. A reference architecture is design guidance; the announcement does not establish it as a generally available turnkey product.
Rank #3
- Form Factor: Plug-in Card
- Cooler Type: Active Cooler
- Maximum Power Consumption: 70W
- Length: 6.6
- Height: 2.7
For a deployment, the rack design is only one part of the decision. Buyers and operators still need to check the specific system’s tray configuration, power delivery, cooling connections and facility requirements, availability, support, and qualification evidence. The 2024 contribution announcement does not settle those supplier-specific questions.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Which partners NVIDIA named
NVIDIA’s technical blog identified the following companies as making Blackwell-based servers: ASRock Rack, ASUS, Dell Technologies, GIGABYTE, Hewlett Packard Enterprise, Hyve, Inventec, MSI, Pegatron, QCT, Supermicro, and Wiwynn. The list documents NVIDIA’s partner mentions at the time; it does not establish current inventory, prices, exact configurations, or independent qualification.
Quick Recap
Sources
- NVIDIA’s October 15, 2024 announcement
- OCP’s description of the contribution
- OCP’s MGX Accelerated Computing Rack and Trays specification listing
- NVIDIA’s Blackwell ecosystem and reference-architecture blog
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
