# Owning a GB300 NVL72 rack

72 Blackwell Ultra GPUs and 36 Grace CPUs in one liquid-cooled rack. Here’s what owning one takes, and when HGX B300 servers make more sense.

## GB300 NVL72 at a glance

- **GPUs:** 72x NVIDIA B300 (Blackwell Ultra) linked as one system
- **CPUs:** 36x NVIDIA Grace
- **GPU memory:** About 21 TB per rack
- **Cooling:** Liquid only
- **Power:** About 130-140 kW per rack
- **Ballpark price:** $8-8.6M per rack, with networking and installation

## Is a rack right for you?

It fits the largest models, where 72 GPUs need to work as one. For most inference, [HGX B300 servers](https://managedgpus.com/b300) use the same GPU, run air-cooled, and can be bought one at a time.

Vera Rubin, NVIDIA’s next generation, comes as racks and as 8-GPU servers. Early supply goes to the largest buyers. Tell us if you want to be in line.

## What it takes to own one

- **A liquid-cooled site.** Few data centers can cool racks this dense, and their space goes fast.
- **Power and floor.** About 130-140 kW per rack, and a floor that can hold it.
- **Lead time.** Orders placed now arrive in about six months.
- **Compliance.** Expect a thorough review of the buyer, the site, and the use.

## We guide. You decide.

1. **Scope it.** Tell us your workload, budget, and timeline. You get a plan and ballpark costs. One call.
2. **We build it.** We compare vendors and data centers side by side. You pick, and we handle the rest in your name. 60-90 days typically.
3. **We run it.** Monitoring, repairs, and warranty claims, on an annual contract. Ongoing.

## What it costs

We take no cut of your hardware or data center. Brokers often earn a percentage of what you spend. We charge flat fees, so a better price only saves you money.

### Paid to vendors, in your name

- **Hardware.** Servers, networking, and shipping. Paid to the vendor, with nothing added.
- **Colocation.** Space, power, and cooling, usually on a 3-year contract. Paid to the data center.

### Paid to us, flat

- **Setup fee.** $25K for 1-7 servers. Half at signing, half when live. Custom pricing for 8 or more.
- **Management fee.** From $1K a month per 4 servers. Covers monitoring, repairs, and warranty claims.

Want a customer for the GPUs? We charge 2.5% of the contract, only if it signs.

Ballpark: one B300 server (8 GPUs)

- **The server:** $550-650K
- **Shipping:** $2,500
- **Data center:** $4-5K a month
- **Our setup fee:** $25K
- **Our management fee:** $1K a month

A cluster also needs networking, about $50-70K per server.

**Don’t want to pay for the hardware all at once?**

- Put in half yourself, and we can usually [introduce a lender](https://gpulenders.com) for the other half.
- Lenders want to see cash or funding of 1.5x the hardware cost. For a $550K server, you put in $275K, the lender covers $275K, and you show $825K.

## Tell us what you’re planning.

- A call with our CTO about what you need
- Ballpark costs and timelines for your budget
- A straight answer if it’s not a fit

Prefer email? [hello@amcompute.com](mailto:hello@amcompute.com)

## NVIDIA GB300 NVL72 racks: questions

**How much does a GB300 NVL72 rack cost?**

Ballpark $8-8.6M per rack, including networking and installation. The data center comes on top.

**Can it be air-cooled?**

No. It needs liquid cooling and a data center built for it.

**How long until delivery?**

Orders placed now arrive in about six months. The largest buyers are served first.

**GB300 NVL72 or HGX B300?**

Same GPU. The NVL72 links 72 of them for the largest workloads. HGX B300 links eight per server and fits far more data centers.

## Further reading

- [Liquid vs air cooling for GPU servers](https://www.amcompute.com/blog/liquid-vs-air-cooling)
- [The power budget of an AI data center](https://www.amcompute.com/blog/the-power-budget-of-an-ai-data-center)
- [HGX, DGX, MGX: NVIDIA’s server platforms](https://www.amcompute.com/blog/hgx-dgx-mgx)

## More hardware and hosting

- [NVIDIA B300 servers](https://managedgpus.com/b300) - Ballpark pricing, power, and lead times for HGX B300 servers.
- [NVIDIA B200 servers](https://managedgpus.com/b200) - Ballpark pricing, power, and availability for HGX B200 servers.
- [NVIDIA H200 servers](https://managedgpus.com/h200) - Ballpark pricing and availability for HGX H200 servers.
- [NVIDIA RTX PRO 6000 servers](https://managedgpus.com/rtx-pro-6000) - Ballpark pricing and power for RTX PRO 6000 servers, the lower-cost way in.
- [GPU colocation](https://managedgpus.com/gpu-colocation) - A U.S. data center for the GPU servers you own.

## Who it’s for

- [AI startups](https://managedgpus.com/ai-startups) - Own your GPU servers instead of renting cloud GPUs.
- [Universities and research labs](https://managedgpus.com/research) - A private GPU cluster in your institution’s name, run for you.
- [Family offices and funds](https://managedgpus.com/family-offices) - Own GPU servers in your own entity and rent them to AI companies.
- [Individual investors](https://managedgpus.com/individual-investors) - Own one or a few GPU servers through your LLC, rented out and run for you.

## Planning a rack? We’ll tell you what’s realistic.

[Plan a deployment](https://managedgpus.com/gb300-nvl72#discuss)

Source: https://managedgpus.com/gb300-nvl72