Vera Rubin NVL72 in Japan: Availability and Expected Pricing

◆ THE SHORT ANSWER

Yes. NVIDIA Vera Rubin NVL72 (VR200) registration is open for Japan through vetted partners, though no Japan-specific deployment has been confirmed yet. Fujitsu's domestic "Made in Japan" manufacturing of B300 servers at its Kasashima plant shows Japan's willingness to build sovereign Blackwell Ultra-generation capacity quickly, a genuine signal for how fast Vera Rubin access could follow once NVIDIA confirms a partner. Vera Rubin NVL72 pairs 72 Rubin GPUs with 36 Vera CPUs (88 Arm Olympus cores each) in a single rack, each GPU carrying 288GB of HBM4 at 22TB/s. The revised APPI applies extraterritorially, so in-country placement removes the consent step for Japanese residents' personal information once capacity lands. For deeper, more established inventory, GB300 and B300 in Japan remain the safer default today.

◆ PRICING

What Vera Rubin capacity is expected to cost in Japan

Vera Rubin pricing in Japan has no established local rate yet. Fujitsu's domestic B300 manufacturing precedent suggests Japan could move quickly once a Vera Rubin partner is confirmed. Register interest and we will confirm what's actually securable once Japanese capacity opens.

Market reference as of September 2026, quoted in USD, not Japan-specific quotes. This is an early, thinly-tracked market globally with no confirmed Japan-specific deployment yet.

Wholesale rates through GPUaaS.com are quoted per enquiry and vary by commitment term, configuration and placement.

$0$2.50$5$7.50$10$12.50$15/GPU-HR

Early-access rate, reservation-only

Global tracked provider, reservation-only

$11.00

Industry reference (3-yr rental TCO)

Per-chip figure, July 2026 analysis

$8.50

Approx. premium band, high end

Directional, thin provider set

~$16.50

Hyperscaler on-demand (projected)

Likely floor once hyperscalers quote broadly

~$18.00

◆ GPUaaS.com wholesale

Vetted partners · direct operator contract

quoted per enquiry

◆ VERY THIN, EARLY MARKET — GLOBAL REFERENCE FIGURES

◆

Where Japanese Vera Rubin capacity will earn its keep

The workloads Japanese Vera Rubin capacity is built for.

Fujitsu's domestic "Made in Japan" manufacturing of B300 servers at its Kasashima plant, launched March 2026 as a response to geopolitical supply chain concerns, shows Japan's genuine willingness to build sovereign Blackwell Ultra-generation capacity quickly. No Vera Rubin-specific manufacturing commitment has been made yet, but the precedent is real. Hokkaido's cheap power should matter even more here than for earlier generations once Vera Rubin's higher draw is deployed.

/01

Next-generation coherent training

Plan ahead for frontier training runs that will need coherent memory across dozens of GPUs at once, on the domestic manufacturing precedent Fujitsu has already set with B300.

Vera Rubin NVL72 · Rubin GPU · 288GB HBM4

/02

Frontier-scale inference

Prepare for multi-trillion-parameter reasoning models in production, where NVIDIA's early figures put Rubin's NVFP4 throughput well beyond GB300's step up from Hopper.

Vera Rubin rental · NVFP4 · 50 PFLOPS

/03

Long-context at rack scale

Hold more context in-rack than any generation covered on this site once available, under the revised APPI's consent exemption for in-country placement.

20.7TB aggregate · 288GB per GPU · APPI Art.24

/04

Agentic workloads, early access

Get ahead of the queue for high-concurrency, multi-agent deployments once Japanese Vera Rubin capacity opens.

Vera Rubin cluster · Dynamo · multi-agent

◆ LIVE NETWORK · 12 LOCATIONS

Vetted GPU partners worldwide, tracking Vera Rubin rollout.

Vera Rubin remains concentrated among eight named cloud partners globally, none confirmed for Japan yet, though Fujitsu's B300 manufacturing precedent is a real signal. Register interest and we will confirm the moment Japanese capacity opens.

Read the full guide to GPU cloud in this location →

GPU GENERATIONS

20+

VETTED PARTNERS

PLACEMENT OPTIONS

24h

QUOTE TURNAROUND

◆ USA◆ CAN◆ UK◆ DEU◆ FRA◆ NLD◆ UAE◆ SAU◆ IND◆ SGP◆ JPN◆ AUS

◆ COST COMPARISON

See how much you save at scale

Wholesale rates against cloud list price for a 64-GPU cluster.

CLUSTER SIZE

8 HGX nodes

64 × GPUS · 730 HRS/MO

ASSUMPTIONS · BLENDED $6.00/GPU-HR · INDICATIVE ONLY

SOURCEEST. MONTHLYVS GPUAAS

Retail cloud

On-demand list price · reserved discounts require lock-in

~$280k

+$84k

Direct datacentre negotiation

Long-term commitment · slow procurement cycle

~$230k

+$34k

◆ BEST VALUE

GPUaaS.com wholesale

Vetted partners · direct operator contract · quotes in 24 hours

~$196k

SAVE ~$84k/MO

Need single-GPU compute? packet.ai has you covered.

◆ HOW IT WORKS

A matchmaker, not a marketplace.

We connect you to our vetted partners. You contract directly with the operator running your nodes.

STEP 01/4

Tell us the requirement

GPU model, count, placement and timeline. Add workload detail if you have it.

STEP 02/4

We match capacity

We find vetted partners with capacity that fits, in the jurisdiction you need.

STEP 03/4

Quotes in 24 hours

Real quotes from partners who hold the capacity, not listings that may not exist.

STEP 04/4

Contract and provision

You contract directly with the operator. We smooth the provisioning process.

Ready to spec your cluster?

Get H100 cluster quotes
in under 24 hours.

Tell us the essentials. We'll line up real quotes from vetted wholesale providers . direct, no platform fee.

◆Quotes in under 24 hours

◆Direct contact with operators

◆No middle-man markup

◆20+ vetted providers · 10 regions

ESSENTIALS

OPTIONAL

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.

◆ FAQ

Frequently Asked Questions

Can I register interest in Vera Rubin NVL72 capacity in Japan?

Yes, through vetted partners, though no Japan-specific deployment has been confirmed yet. Fujitsu's domestic B300 manufacturing at Kasashima is a genuine signal for how quickly Japan could move on Vera Rubin. Register your interest and we will confirm what's actually securable once Japanese capacity opens.

How much will Vera Rubin cost per hour in Japan?

Tracked early-access pricing globally starts around $11.00/hr on a reservation-only basis, with no Japan-specific rate established yet given the market's early stage here.

What are the Vera Rubin NVL72 specs?

Full Vera Rubin NVL72 specs: 72 Rubin GPUs paired with 36 Vera CPUs (88 Arm Olympus cores each), 288GB HBM4 per GPU at 22TB/s bandwidth, connected via NVLink 6. A rack draws an estimated 190-230kW, meaningfully more than GB300's 135-140kW. Some providers brand the underlying Rubin GPU as "H300"; it's the same R100 silicon.

Does Fujitsu's domestic B300 manufacturing suggest early Vera Rubin access?

It's a genuine precedent. Fujitsu's "Made in Japan" sovereign AI servers with HGX B300 GPUs, manufactured domestically at Kasashima from March 2026, show Japan can move quickly on Blackwell Ultra-generation hardware, though no Vera Rubin-specific commitment has been made yet.

Should I wait for Vera Rubin instead of renting B300 in Japan now?

B300 is available now in Japan through Fujitsu's domestic manufacturing and specialist providers. Vera Rubin remains genuinely thin globally with no confirmed Japan date. If your workload needs capacity this year, B300 is the more realistic choice.

Will Hokkaido siting matter more for Vera Rubin than earlier generations?

Likely, once deployed. Vera Rubin's estimated 190-230kW per rack draw is well above B300's, so the power saving Hokkaido siting already offers for earlier generations should translate into a larger absolute saving here.

Vera Rubin in Japan, at wholesale price.