# Own your GPUs instead of renting them

B300, B200, H200, or RTX PRO 6000 servers in your company’s name, hosted in a U.S. data center and run for you.

## When owning beats renting

- **Steady usage.** If your GPUs stay busy for two years or more, owning usually costs less than renting.
- **A predictable cost.** A one-time purchase and a fixed data center contract. No surprise rate changes.
- **Dedicated hardware.** No one else runs on your servers, which makes customer security reviews easier.
- **Resale value.** When you upgrade, you can sell or rent out the old servers.

## When to keep renting

If your usage is spiky or you only need GPUs for a few months, keep renting. We’ll tell you on the first call if owning doesn’t make sense yet. Compare both for your setup in the [rent vs own calculator](https://managedgpus.com/rent-vs-own), or read [the math on owning vs renting](https://www.amcompute.com/blog/tco-own-vs-rent-gpu-clusters).

## Which GPU for which job

**B300**

288 GB per GPU, for large models and training. Ballpark $550-650K per server. [More on B300](https://managedgpus.com/b300)

**B200**

180 GB per GPU, good value for inference. Ballpark $450-500K. [More on B200](https://managedgpus.com/b200)

**H200**

141 GB per GPU, lower power and easy to get. Ballpark $340K. [More on H200](https://managedgpus.com/h200)

**RTX PRO 6000**

96 GB per GPU, for inference and fine-tuning at the lowest cost. Ballpark $100-150K. [More on RTX PRO 6000](https://managedgpus.com/rtx-pro-6000)

## 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 run today.

- 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)

## Questions from startups

**Are we too early to buy GPUs?**

If your usage is small or changes every month, probably. Owning pays off when the servers stay busy.

**Do we pay everything upfront?**

Vendors usually take a 20-50% deposit and the rest before shipping. Lenders can finance part of it, and a signed customer contract makes that much easier.

**How many servers should we start with?**

One is enough. Many startups start with 8 to 16. Our setup fee is the same for 1 to 7.

**Can we rent out capacity we don’t use?**

Yes. We can find a customer on a multi-year contract, for 2.5% of the contract if it signs.

**Who has access to our servers?**

Your team holds admin access. We manage the hardware, not your data or models.

## Further reading

- [Owning vs renting GPU clusters](https://www.amcompute.com/blog/tco-own-vs-rent-gpu-clusters)
- [How to finance a GPU cluster](https://www.amcompute.com/blog/how-to-finance-a-gpu-cluster)
- [GPU financing (GPU Lenders)](https://gpulenders.com)

## 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.
- [NVIDIA GB300 NVL72 racks](https://managedgpus.com/gb300-nvl72) - What it takes to own a liquid-cooled GB300 NVL72 rack.
- [GPU colocation](https://managedgpus.com/gpu-colocation) - A U.S. data center for the GPU servers you own.

## Ready to stop renting? We’ll run it.

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

Source: https://managedgpus.com/ai-startups