Rackscale AI Accelerators

Fully co-designed rack- and POD-scale AI systems that combine accelerators, host CPUs, scale-up fabric (NVLink-class domains), and scale-out networking in one thermal/mechanical package. Lineage runs from multi-GPU HGX/DGX boxes through Grace Hopper NVL32 domains to liquid-cooled 72-GPU racks (GB200, GB300, Vera Rubin) and competing racks such as AMD Helios. NVIDIA, AMD, and OEM rack builders (Dell, Supermicro) ship the current generation for AI-factory halls.

Hidden Factory

Rackscale AI Accelerators

2023 → 2026
Swipe roadmap
Exponential Industry Backdrop
milestone
May 28, 2023
NVIDIA DGX GH200 AI supercomputer announcement
NVIDIA Corporation
milestone
Mar 16, 2026
NVIDIA Vera Rubin platform announcement
NVIDIA CorporationFirst production NVIDIA Vera Rubin systems arrive at Microsoft datacentersNVIDIA Blackwell platform and GB200 NVL72 announcement (GTC 2024)
milestone
Jul 23, 2026
AMD Helios rack-scale AI launch
Advanced Micro Devices
partnership
Aug 10, 2026
NVIDIA and six institutions launch AI-factory financing platforms (>$500B)
NVIDIA CorporationApolloBlackRock
milestone
Aug 18, 2026
Etched delivers first inference rack to Jane Street
Etched
milestone
Aug 21, 2026
First production NVIDIA Vera Rubin systems arrive at Microsoft datacenters
Microsoft CorporationNVIDIA CorporationSatya Nadella
partnership
Aug 26, 2026
NVIDIA Groq 3 LPX Enters Full Production; Nebius Deploys in Token Factory
NVIDIA CorporationNebius
2026
partnership

NVIDIA Groq 3 LPX Enters Full Production; Nebius Deploys in Token Factory

NVIDIA announced full production of the rack-scale Groq 3 LPX system (256 LPUs, 128GB SRAM, 640TB/s bandwidth), with Nebius deploying clusters into the Nebius Token Factory for low-latency agentic decoding achieving 3,400 tok/s on Gemma 4 31B /NVIDIA Newsroom/ /Nebius/.

Related: NVIDIA Corporation · organization Nebius · organization
partnership

SpaceXAI adopts NVIDIA Vera CPU and Vera Rubin for Agentic AI

SpaceXAI announced deployment of NVIDIA Vera CPUs featuring 88 Olympus cores and 1.2TB/s memory bandwidth to accelerate orchestration and code execution for Grok agentic AI workloads /NVIDIA Newsroom/ /X (formerly Twitter)/.

milestone

First production NVIDIA Vera Rubin systems arrive at Microsoft datacenters

Satya Nadella (Aug 21, 2026) posted that it was delivery day at Microsoft datacenters as the first production Vera Rubins arrived, thanking NVIDIA and Azure hardware and datacenter teams /X/.

Related: Microsoft Corporation · organization NVIDIA Corporation · organization Satya Nadella · person
milestone

SK hynix publishes CPO roadmap in Nature Electronics

SK hynix newsroom Aug 20, 2026: paper "Co-packaged optics for high-performance computing and artificial intelligence" in Nature Electronics (s41928-026-01681-6), corresponding authors Seunghoon Hong (SK hynix) and Kyusang Lee (UVA). Targets >100 Tb/s per node, <1 pJ/bit, <10 ns latency; optics-centric photonic interposer between XPU and memory pools to break the rack/pod bandwidth wall /SK hynix/ /Nature Electronics/.

milestone

Etched delivers first inference rack to Jane Street

Etched PR Aug 18, 2026: first rack shipped last month; Jane Street said it tested the chip, is pleased with early results, and has its own rack running in its data center. Treated as the announcement day; ship month is July 2026 without a calendar day /GlobeNewswire/.

Related: Etched · organization
partnership

NVIDIA and six institutions launch AI-factory financing platforms (>$500B)

NVIDIA announced MOUs with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to set up independent compute-financing platforms designed to mobilize over $500 billion of third-party capital for AI-factory buildout. NVIDIA said the figure is not NVIDIA revenue, a single fund, or one customer commitment; partners independently underwrite each deal. Jensen Huang said NVIDIA may offer residual-value support on up to 25% of an opportunity. Final agreements were still outstanding /NVIDIA/ /Jensen Huang / X/.

Related: NVIDIA Corporation · organization Apollo · organization BlackRock · organization
milestone

AMD Helios rack-scale AI launch

AMD launches Helios rack-scale AI infrastructure (72× MI455X) at Advancing AI 2026 (AMD blog).

Related: Advanced Micro Devices · organization
milestone

CoreWeave measures Vera Rubin NVL72 10x tokens per megawatt vs GB200

CoreWeave published first measured silicon performance on Vera Rubin NVL72. At a matched interactivity target, the rack generated 10x tokens-per-second per megawatt versus NVIDIA GB200 NVL72 on the same DeepSeek R1 workload, using TensorRT-LLM and Dynamo with NVFP4, expert parallelism, multi-token prediction, and disaggregated prefill/decode /CoreWeave/.

Related: CoreWeave · organization NVIDIA Corporation · organization
Video explainer
CoreWeave AI Data Centers: Closed-Loop Cooling at Scale
Click to load video
CoreWeave AI Data Centers: Closed-Loop Cooling at Scale Watch ↗
partnership

First Dell PowerEdge XE9812 Vera Rubin NVL72 delivery to CoreWeave

Michael Dell announced delivery of the world's first working liquid-cooled Dell PowerEdge XE9812 NVIDIA Vera Rubin NVL72 server rack for CoreWeave (X, May 31, 2026).

Related: Dell Technologies · organization CoreWeave · organization NVIDIA Corporation · organization
milestone

NVIDIA Vera Rubin platform announcement

NVIDIA announces Vera Rubin platform and related racks at GTC; seven chips in full production.

Related: NVIDIA Corporation · organization First production NVIDIA Vera Rubin systems arrive at Microsoft datacenters · process NVIDIA Blackwell platform and GB200 NVL72 announcement (GTC 2024) · process
2024
milestone

NVIDIA Blackwell platform and GB200 NVL72 announcement (GTC 2024)

GTC 2024 (Mar 18) launch of the NVIDIA Blackwell platform: Blackwell GPU architecture, GB200 Grace Blackwell Superchip, liquid-cooled GB200 NVL72 rack-scale system (72 Blackwell GPUs + 36 Grace CPUs on fifth-gen NVLink), HGX B200, sixth-generation air-cooled DGX B200 (8 Blackwell GPUs, two 5th Gen Intel Xeon; up to 144 PFLOPS FP4), and broad cloud/OEM adoption commitments. Foundational rackscale-AI milestone that established the NVL72 rack product line later continued by GB300 and Vera Rubin NVL72; DGX B200 is the traditional air-cooled 8-GPU SuperPOD/BasePOD building block, not NVL72 /NVIDIA Newsroom/.

Related: NVIDIA Corporation · organization Jensen Huang · person NVIDIA Vera Rubin platform announcement · process
2023
milestone

NVIDIA DGX GH200 AI supercomputer announcement

DGX GH200 large-memory AI supercomputer class with Grace Hopper Superchips and NVLink Switch System—predecessor NVLink domain architecture before Blackwell NVL72.

Related: NVIDIA Corporation · organization