GB300 Coming soon. Learn more →

Instance Pricing & Calculator

A breakdown of instance, storage, and networking costs.

Contact

Gpu Instances & Clusters

On-demand pricing

Select Instance

  • H200

    144GB SXM
    128 vCPU
    2TB RAM
    30.7TB NVME
    3200TBPS RDMA

  • H100

    80GB SXM
    128 vCPU
    2TB RAM
    30.7TB NVME
    3200TBPS RDMA

  • L40S

    48GB PCIE
    64 vCPU
    2TB RAM
    30.4TB NVME
    1600TBPS RDMA

  • A100

    80GB SXM
    256 vCPU
    1TB RAM
    30.7TB NVME
    1600TBPS RDMA

  • A40

GPUs per instance
Term

For committed and dedicated access with long term pricing, contact sales.

Customize your plan from the choices available.

CPU Instances

Powerful AMD-based CPU instances, scaled to match your needs

Powerful AMD-based CPU instances, scaled to match your needs. Configurable to attach to any GPU cluster as part of the same network fabric. For CPU instances (v) is a virtualized type, and (m) denotes metal.

Instance Name
vCPUs
Memory
Storage
Maximum Bandwidth
Monthly Price
c1.4.v
4
32 GB
100 GB
3 Gbps
$73
c1.8.v
8
64 GB
200 GB
6 Gbps
$146
c1.16.v
16
128 GB
400 GB
12 Gbps
$292
c1.32.v
32
256 GB
800 GB
25 Gbps
$584
c1.64.v
64
512 GB
1,600 GB
50 Gbps
$1,168
c1.128.v
128
1,024 GB
3,200 GB
100 Gbps
$2,336
c1.256.m
256
2,048 GB
6,400 GB
200 Gbps
$4,672

Start training, testing, or deploying today with Firmus AI Cloud.

Storage

Fast NVMe onboard storage is complemented by ultra-performant parallel file storage from WEKA and an S3-compatible object file store. Each node or instance in a Firmus cluster is connected via a dedicated storage fabric at up to 200 GB/s, non-blocking.

WEKA NVME PFS
$0.11 per GB / month

Networking

RDMA InfiniBand east-west compute fabric, capable of up to 3,200 GB/s, comes standard with Firmus clusters, along with a dedicated, non-blocking in-band and storage network. For public connectivity, Firmus offers flexible ways to access clusters to end users.

FAQs

Do you have H100 capacity?

Yes. Firmus has cloud regions built or under build in locations around the world.

All AZs in these regions feature our high-efficiency hosting infrastructure.

Each region is capable of hosting multi-thousand single-node clusters within the same network fabric, based on customer requirements.

H100 & GB300 infrastructure is being installed in these regions; however, with current market dynamics, demand is high, and so we work closely with customers to capacity plan for current and current and future infrastructure.

Do you offer custom pricing?

Yes. Our sales team can help craft a solution with pricing reflective of custom contract terms and payment terms - including up-front payment amounts, which can further reduce the quoted rate. We're always ready to work with partners who show a strong commitment to growing with Firmus and are ready to structure solutions that work for everyone.

How does billing work?

Approved and on-boarded accounts receive bills for consumption based on contracted capacity.

Is connectivity included?

A minimum 1gbps/1gbps IP transit is provided with each cluster. For higher capacity ingress and egress, customized connectivity solutions are available, which may incur additional costs, depending on the engineered solution.

In addition to IP transit, Firmus provides a range of options to cross-connect into the public cloud.

How can I sign up?

The first step to coming on board is filling out our 'Enquire Now' form with information about you and your needs. Within a day or two, a team member will be in contact with you to arrange a scoping call.

Optimized for performance and scale

Start training, testing, or deploying today with Firmus.

Get in touch