> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.ibee.ai/docs/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.ibee.ai/docs/_mcp/server.

# Create a GPU VM

> Deploy a GPU-enabled virtual machine on IBEE Solutions for AI, ML, rendering, and compute-intensive workloads.

Deploy a GPU-enabled virtual machine (VM) from the IBEE Solutions console. GPU VMs include dedicated NVIDIA GPUs for AI/ML training, inference, rendering, and other GPU-accelerated workloads.

## Before you begin

* An active IBEE Solutions account
* A verified account to unlock server deployment ([Verify your account](/docs/getting-started/account-setup/verify-account))
* Sufficient wallet credits — the platform checks your balance before deploying ([Billing and usage](/docs/platform-fundamentals/billing-and-usage))

## Deploy a GPU VM

### 1. Open GPU VMs

In the portal sidebar, click **GPU VMs** under **Products**, then click **Create server**.

### 2. Select Region and Location

Select a region tab and choose a data center location within it. GPU availability varies by location.

### 3. Select a GPU Plan

Choose a plan based on your GPU, compute, and memory requirements. Plans include dedicated NVIDIA GPUs with varying VRAM, vCPUs, and system RAM. The estimated monthly and hourly cost updates as you select.

### 4. Select OS

Pick a source for the boot disk under **Operating System**:

| Tab           | Use it for                                               |
| ------------- | -------------------------------------------------------- |
| **Templates** | GPU-optimized OS images maintained by IBEE. The default. |
| **ISOs**      | Boot from a public or uploaded ISO.                      |
| **Snapshots** | Restore a manual recovery point from another GPU VM.     |
| **Backups**   | Restore from a scheduled backup recovery point.          |

The available OS templates are filtered to show only images compatible with GPU VMs — these include pre-configured NVIDIA driver support.

### 5. Select OS Version

Choose a version from the **OS Version** dropdown.

### 6. Enter Hostname

Enter a unique, lowercase hostname for your server (letters, numbers, hyphens only).

### 7. Deploy

Click **Deploy Now**. The platform first checks your organization's billing state and wallet balance. If credits are sufficient, your GPU VM begins provisioning. If not, a **Billing needs attention** popup appears — click **Add Credits** to top up and return.

## Verify

Once provisioned, your GPU VM appears in the GPU VMs list with a running status. Click it to open the detail page, which shows the GPU configuration, SSH key names, and assigned IP.

Confirm GPU access inside the VM:

```bash
nvidia-smi
```

## Manage your GPU VM

From the server detail page, use **Actions** to control your server:

| Action | Description                           |
| ------ | ------------------------------------- |
| Start  | Power on a stopped server             |
| Stop   | Gracefully shut down a running server |
| Reboot | Restart the server                    |
| Delete | Permanently delete — cannot be undone |

> **Warning**
>
> Deleting a GPU VM is permanent and cannot be undone. Back up your data and snapshots before deleting.

## Troubleshooting

**Deploy Now button is disabled**
Ensure all required fields are filled — Region, Location, Plan, OS, OS Version, and Hostname.

**Billing needs attention popup**
Your organization's wallet balance is insufficient. Click **Add Credits** to top up, then return.

**`nvidia-smi` not found**
The NVIDIA driver may not be installed on the selected template. See [CUDA and driver setup](/docs/infrastructure/gpu-vms/cuda-and-driver-setup).

**Server stuck in provisioning**
Wait a few minutes. If the status does not update, contact IBEE support.

## Related pages

* [GPU types](/docs/infrastructure/gpu-vms/gpu-types)
* [CUDA and driver setup](/docs/infrastructure/gpu-vms/cuda-and-driver-setup)
* [AI and ML workloads](/docs/infrastructure/gpu-vms/ai-and-ml-workloads)
* [SSH Keys](/docs/tools/ssh-keys)