# Welcome to Denvr AI Cloud Docs!

<table data-view="cards"><thead><tr><th></th><th></th><th></th></tr></thead><tbody><tr><td><a href="overview/data-centers">AI clusters</a></td><td><p></p><p>Data center GPUs including NVIDIA H100, A100 &#x26; A40 with compute fabric up to 3.2Tbps of non-blocking InfiniBand.  </p></td><td></td></tr><tr><td><a href="platform/storage">Storage</a></td><td><p></p><p>High-performance direct attached and network filesystems to cache and persist data for your compute.</p></td><td></td></tr><tr><td><a href="api-reference">API reference</a></td><td><p></p><p>Automate provisioning and management of applications, virtual machines, and bare metal hosts.</p></td><td> </td></tr></tbody></table>

Users can provision AI infrastructure in three primary ways:

* [Applications](https://docs.denvrdata.com/docs/platform/applications) - managed Notebooks, IDEs, and private containers running on bare metal
* [Virtual machines](https://docs.denvrdata.com/docs/overview/getting-started/launch-a-virtual-machine) - support for 1x, 2x, 4x, and 8x GPU configurations
* [Bare metal ](https://docs.denvrdata.com/docs/platform/bare-metal)- direct access to host for peak performance