> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/confidential-computing/_mcp/server.

# Deploy Confidential Virtual Machines

> Configure a host, a confidential virtual machine, and NVIDIA GPUs for Confidential Computing.

This guide provides the steps to configure a system for Confidential Computing with NVIDIA Hopper and Blackwell GPUs, from the BIOS through GPU attestation in the guest.
For the trusted components, threats, and protection limits behind these steps, refer to [Trust and Threat Model](/confidential-computing/virtual-machines/threat-model).

Complete the following sections in order, or start with the section for your [persona](/confidential-computing/virtual-machines#personas):

1. [Configuring the Hardware](/confidential-computing/virtual-machines/deployment-guide/hardware): Configure the BIOS for TDX or SEV-SNP.
2. [Configuring the Host](/confidential-computing/virtual-machines/deployment-guide/host): Prepare the host OS and the hypervisor.
3. [Configuring the GPUs](/confidential-computing/virtual-machines/deployment-guide/gpu-mode): Bind the GPUs to VFIO and set the GPU mode.
4. [Creating a Confidential VM](/confidential-computing/virtual-machines/deployment-guide/create-vm): Create and launch a CVM with the GPUs attached. Skip this section if you already have a VM orchestration flow.
5. [Configuring the Guest](/confidential-computing/virtual-machines/deployment-guide/guest): Install and configure the NVIDIA driver in the CVM.
6. [Attesting the GPU](/confidential-computing/virtual-machines/deployment-guide/attest): Verify the configuration, attest the GPU, and enable the GPU to accept work.

## Conventions

In the command examples, lines that start with `$` are commands to run, and lines that start with `#` are comments.
Lines without a prefix are command output.
Some examples omit output that doesn't matter unless the command reports an error.

The host examples store downloads and scripts in the `/shared` directory.
You can use a different directory if you change the paths in the commands.

> Configure a host, a confidential virtual machine, and NVIDIA GPUs for Confidential Computing.

## Docs

- [Configuring the Hardware](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/hardware.md): Select CPUs and GPUs for Confidential Computing and configure the BIOS for AMD SEV-SNP or Intel TDX.
- [Configuring the Host](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/host.md): Prepare an AMD SEV-SNP or Intel TDX host and its hypervisor to run confidential virtual machines with NVIDIA GPUs.
- [Configuring the GPUs](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/gpu-mode.md): Bind NVIDIA GPUs to VFIO and set the Confidential Computing or Protected PCIe mode before you pass the GPUs to a confidential virtual machine.
- [Creating a Confidential VM](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/create-vm.md): Create and launch a confidential virtual machine with NVIDIA GPUs attached on an Intel TDX or AMD SEV-SNP host by using KVM and QEMU.
- [Configuring the Guest](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/guest.md): Install and configure the NVIDIA driver in a confidential virtual machine and verify the GPU Confidential Computing state.
- [Attesting the GPU](https://docs.nvidia.com/confidential-computing/virtual-machines/dev/deployment-guide/attest.md): Verify the Confidential Computing configuration, attest the GPU, and set the GPU ready state in a confidential virtual machine.