AI Infrastructure

NVIDIA Blackwell Platform

The data-center GPU platform most frontier AI models are trained and served on.

Nobody buys a Blackwell chip at retail — entry means a seven-figure rack or renting by the GPU-hour. NVIDIA's platform spans B200 GPUs, GB200 superchips pairing two GPUs with a Grace CPU, and the GB200 NVL72: 72 GPUs and 36 Grace CPUs fused into one liquid-cooled rack that behaves like a single accelerator. The Blackwell Ultra refresh (GB300, with 288GB of HBM3e per GPU) now carries volume shipments, with the Vera Rubin generation queued behind it.

Best for

Hyperscalers, frontier labs, and enterprises provisioning training or high-throughput inference clusters — and, realistically, anyone renting Blackwell capacity through CoreWeave, AWS, Azure, or Google Cloud rather than buying racks.

Against AMD's Instinct MI300/MI350 line and Google's TPUs, Blackwell's real moat is NVLink rack-scale coherence plus CUDA: the NVL72 makes 72 GPUs address one memory fabric, and two decades of CUDA software mean every framework is tuned for it first. AMD now competes credibly on memory capacity and price per GPU; nobody yet matches the full rack-as-one-computer package.

Core features

AI compute

B200 and Blackwell Ultra GB300 GPUs built for frontier training and low-precision inference, with up to 288GB of HBM3e per GPU on the Ultra refresh.

Platform ecosystem

One stack from silicon to software: Grace CPUs, NVLink, InfiniBand and Spectrum-X networking, and the CUDA libraries every major AI framework targets.

Data-center scale

GB200 and GB300 NVL72 racks fuse 72 GPUs into a single coherent domain — the unit hyperscalers now plan AI factories around.

Pros

+ NVL72 rack-scale design makes 72 GPUs behave as one accelerator — no rival ships an equivalent turnkey unit
+ CUDA plus NVLink and InfiniBand networking mean every major framework and model is optimized for Blackwell first
+ The Blackwell Ultra (GB300) refresh extends the platform to 288GB of HBM3e per GPU without a rack redesign

Watchouts

- Entry costs run $2–3 million per NVL72 rack, plus liquid cooling and power densities most existing data centers can't supply
- The announced Vera Rubin successor generation puts a depreciation clock on every Blackwell purchase

Power List links

People behind NVIDIA Blackwell Platform

Full list

Jensen HuangNVIDIA

Pricing

enterprise

No list price — B200 cited at $30K–$40K; NVL72 racks ~$2M–$3M

NVIDIA publishes no list prices; systems sell through OEMs and clouds. Jensen Huang put B200 at $30,000–$40,000; GB200 superchips are reported at $60,000–$70,000 and full NVL72 racks at $2–3 million. Cloud rental of GB200 capacity is reported from around $10.50 per GPU-hour at specialist providers.

Full NVIDIA Blackwell Platform pricing breakdown →

Discovery tags

AI Infrastructure Data AnalysisCompute Layer

Company context

NVIDIA sells the GPUs, systems, networking, and CUDA software that most AI training and inference still runs on.

NVIDIA Stock Page

TECHi verdict

A 95 here is not a consumer recommendation — you cannot buy Blackwell the way you buy a laptop — it's a statement about market position. Blackwell is the compute bottleneck the AI buildout runs through: when hyperscalers and frontier labs plan capacity, they plan in NVL72 racks. The honest caveats: AMD's MI350 generation competes on memory capacity and price, every hyperscaler is shipping in-house silicon to escape NVIDIA's margin, and Rubin's arrival puts a date stamp on every Blackwell purchase. For most enterprises the practical call is renting Blackwell capacity from a cloud rather than owning racks — let someone else eat the depreciation on a platform NVIDIA itself now replaces on a roughly annual cadence.

Related in the directory

People

The people behind NVIDIA

Jensen HuangFounder and CEO, NVIDIA

Maker

Company

NVIDIANVDANVIDIA sells the GPUs, systems, networking, and CUDA software that most AI training and inference still runs on.

Same category

Alternatives in the same category

Sources

Official product page

Alternatives

NVIDIA AI Enterprise

NVIDIA's supported software stack for running production AI on its GPUs.

Strategic Product Influence — Editorial 87/100 · version not recorded

Score methodology

Frequently asked

What is NVIDIA Blackwell Platform?

The data-center GPU platform most frontier AI models are trained and served on.

Who builds NVIDIA Blackwell Platform?

NVIDIA Blackwell Platform is built by NVIDIA, a company with listed shares (NVDA).

How much does NVIDIA Blackwell Platform cost?

NVIDIA Blackwell Platform uses enterprise. Current tiers: No list price — B200 cited at $30K–$40K; NVL72 racks ~$2M–$3M. Check the official site before buying — AI pricing changes often.