One rate card.
Public. Flat.

No sales-only tiers, no secret discounts, no "it depends." Reserve longer, pay less. Egress is free. Storage is $0.04/GB·mo.

Hourly rates, by GPU.

GPU · instance VRAM vCPU RAM Fabric Rate / GPU·hr vs. hyperscaler

* Hyperscaler deltas compare on-demand rates as published by AWS, GCP, Azure as of Q2 2026. Rates include NVLink, local NVMe, and fabric.

Everything else is free.

— inbound
$0.00
Data transfer in, from anywhere.
— outbound
$0.00
Egress to internet, other clouds, end users.
— fabric
$0.00
InfiniBand NDR, NVLink, intra-region traffic.
— support
$0.00
Shared Slack with our platform team. 24/7 for annual.
SSD (NVMe)$0.04 / GB·mo
Object storage$0.015 / GB·mo
Snapshots$0.008 / GB·mo
CPU instancesfrom $0.018 / vCPU·hr
Load balancer$0.012 / hr
Elastic IP$0.00
Private fabriccustom
Dedicated clustercustom
HIPAA / FedRAMPcontact

Pick how you want to buy capacity.

Hourly, no commit

Spin up, run, tear down. Billed per second after the first minute. Perfect for experiments, CI, and spiky inference.

$3.49 / B200·hr
Start →
MOST POPULAR

Reserve by the day

Lock capacity for a training run. As short as 24 hours. 18% off on-demand. No quarterly contract.

$2.86 / B200·hr
Reserve →

12-month commit

Deep discount for teams with sustained demand. Dedicated cluster, private fabric, named SRE.

$1.92 / B200·hr
Talk to us →

Pricing, answered.

How do you price so much lower than AWS or GCP?

We run our own datacenters, buy GPUs directly, and don't carry the overhead of selling 300 other services. We also don't charge for egress or inter-node traffic, which is how hyperscaler bills balloon.

Is there a minimum commit?

No. On-demand clusters bill per second after a 60-second minimum. Reserved starts at 24 hours.

What happens if capacity isn't available?

Our availability page is live. If a region shows green, it's yours. We overprovision rather than oversubscribe — if we can't fulfill a reservation, it's free.

Do you offer academic pricing?

Yes — 40% off on-demand for verified .edu teams, plus shared Slurm clusters with fair-share scheduling. Email research@neoscale.ai.

Can I bring my existing k8s cluster?

Yes. Install our operator; it provisions managed node pools on Neoscale GPUs and streams them into your existing control plane.

What regions are available?

Dallas (DFW-01), Reno (RNO-01), Oakland (OAK-01), and Amsterdam (AMS-01). Phoenix and Singapore come online Q3.

Need something we didn't list?

Contact sales → Open calculator