# Billing

## On-demand pricing

All on-demand instances are billed in one-second increments with pricing quoted per hour.  Metering begins when an instance or application in online until it is terminated.

{% hint style="info" %}
Reserved instances offer significant savings over on-demand due to longer term commitments, pre-payment options, and order size.
{% endhint %}

{% tabs %}
{% tab title="GPU" %}

<table data-full-width="true"><thead><tr><th>Platform</th><th>Instance</th><th width="134">Interconnect</th><th width="140">Price/Hour</th></tr></thead><tbody><tr><td>1x NVIDIA GH200</td><td>72 vCPU, 480GB RAM, 7.6TB SSD</td><td>IB 400G</td><td>$3.87 / GPU</td></tr><tr><td>8x H100 80GB SXM</td><td>208 vCPUs, 1TB RAM, 23TB SSD</td><td>IB 3200G</td><td>$2.10 / GPU</td></tr><tr><td>8x A100 80GB SXM</td><td>208 vCPUs, 1TB RAM, 23TB SSD</td><td> IB 1600G</td><td>$1.30 / GPU</td></tr><tr><td>8x Gaudi2 96GB</td><td>160 vCPUs, 1TB RAM, 30TB SSD</td><td>RoCe 2400G</td><td>$1.25 / GPU</td></tr><tr><td>8x A100 40GB SXM</td><td>128 vCPUs, 1TB RAM, 15TB SSD</td><td>800B IB</td><td>$1.15 / GPU</td></tr><tr><td>4x A100 40GB PCIe</td><td>64 vCPUs, 512GB RAM, 7.6TB SSD</td><td>-</td><td>$1.15 / GPU</td></tr><tr><td>4x A40 48GB PCIe</td><td>128 vCPU, 512GB RAM, 7.6TB SSD</td><td>-</td><td>$0.65 / GPU</td></tr></tbody></table>

Prices are displayed per GPU and include vCPU, system memory and direct-attached NVMe SSD disks.  GPU fabrics (InfiniBand and RoCE V2) are available only in full-node instances.\
\
Virtualization configurations are available with 1x, 2x, 4x and 8x GPUs.
{% endtab %}

{% tab title="CPU" %}

<table><thead><tr><th width="272">Platform</th><th width="214">Instance</th><th width="180">Price/Hour</th></tr></thead><tbody><tr><td>Intel Xeon Gen3 (Ice Lake)</td><td>8 vCPU, 32GB RAM</td><td>$0.31</td></tr><tr><td>Intel Xeon Gen3 (Ice Lake)</td><td>16 vCPU, 64GB RAM</td><td>$0.62</td></tr><tr><td>Intel Xeon Gen3 (Ice Lake)</td><td>32 vCPU, 128GB RAM</td><td>$1.24</td></tr><tr><td>AMD EPYC 3rd Gen (Milan)</td><td>8 vCPU, 32GB RAM</td><td>$0.40</td></tr><tr><td>AMD EPYC 3rd Gen (Milan)</td><td>16 vCPU, 64GB RAM</td><td>$0.79</td></tr><tr><td>AMD EPYC 3rd Gen (Milan)</td><td>32 vCPU, 128GB RAM</td><td>$1.59</td></tr></tbody></table>
{% endtab %}
{% endtabs %}

## Storage pricing

File systems are billed per GB used in hourly increments.  Storage is billed regardless if compute resources are online.

<table><thead><tr><th width="187">Storage Type</th><th width="397">Performance Class</th><th>Price/GB/mo</th></tr></thead><tbody><tr><td>Block</td><td>High-Performance (NVMe SSD)</td><td>$0.08</td></tr><tr><td>File systems</td><td>High-Performance (NVMe SSD)</td><td>$0.08</td></tr><tr><td></td><td>General Purpose</td><td>$0.03</td></tr><tr><td>VM machine images</td><td>High-Performance (NVMe SSD)</td><td>$0.08</td></tr><tr><td>Snapshots</td><td>High-Performance (NVMe SSD)</td><td>$0.08</td></tr></tbody></table>

## Network pricing

Denvr AI Cloud does not currently bill for the following network services:

* Internet ingress/egress
* Public IPs
* VPCs


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.denvrdata.com/docs/platform/billing.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.