> For the complete documentation index, see [llms.txt](https://nx.docs.scailable.net/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://nx.docs.scailable.net/ai-manager-v6.1.4/ai-accelerators-support/intel-support.md).

# Intel Support

| Support Level | AI Accelerator | CPU Architecture | Operating System       | API/driver version |
| ------------- | -------------- | ---------------- | ---------------------- | ------------------ |
| Intel CPU     | Full support   | x86\_64          | Ubuntu 22+ and Windows | -                  |
| Intel GPU     | Experimental   | x86\_64          | Ubuntu 22+ and Windows | -                  |
| Intel NPU     | Experimental   | x86\_64          | Ubuntu 22+ only        | -                  |

To use OpenVINO runtimes, select the option during plugin install on compatible hardware.\
\
If you want to preselect GPU or NPU runtimes, make sure you have the latest Intel drivers installed on your machine.

## Hardware Requirements

Check the [Intel system requirements](https://docs.openvino.ai/2025/about-openvino/release-notes-openvino/system-requirements.html) tables to confirm compatibility.

## Software Requirements: for Ubuntu

For the Intel drivers, follow:\
\
\- For GPU: <https://github.com/intel/compute-runtime/releases/tag/25.09.32961.5>\
\- For NPU: <https://github.com/intel/linux-npu-driver/releases/tag/v1.16.0>

Also make sure you have the latest libtbb installed:

```bash
sudo apt install libtbb12 ocl-icd-libopencl1
```

Add the user to `video` and `render` groups:

```bash
sudo usermod -aG video $USER
sudo usermod -aG render $USER
```

If Ubuntu or Debian overwrites the packages installed above, prevent APT from upgrading them by placing the relevant packages on hold, for instance:

```bash
sudo apt-mark hold intel-ocloc intel-opencl-icd libigdgmm12 libze-intel-gpu1
```

## Software Requirements: for Windows

To make use of the Intel GPU, make sure to install the Intel Graphics Driver as shown on this [page](https://www.intel.com/content/www/us/en/support/articles/000005629/graphics/processor-graphics.html).

## Configuration

By default, the OpenVINO runtime performs model inference on CPU.

To run a model on a compatible Intel GPU, specify that in the JSON file located at `/opt/networkoptix-metavms/mediaserver/var/nx_ai_manager/nxai_manager/bin/runtime_args.json` :

```json
{
  "device_type": "GPU"
}
```

Similarly, to use the Intel NPU, the JSON file should contain:

```json
{
  "device_type": "NPU"
}
```

After making the changes, restart the Mediaserver on [Linux](/ai-manager-v6.1.4/support-and-troubleshooting/troubleshooting/controlling-the-server-and-the-plugin.md#os-linux) or [Windows](/ai-manager-v6.1.4/support-and-troubleshooting/troubleshooting/controlling-the-server-and-the-plugin.md#os-windows).

## Verifying the Runtime Is Available

After installing the Intel drivers and restarting the Mediaserver, the Intel runtime appears in the AI Manager plugin's runtime selection panel. Open any camera's settings, go to the **Integrations** tab, click **AI Manager**, and confirm that the Intel runtime entry is visible.

<figure><img src="/files/B2SF8eeTAaejiahTvb45" alt="AI Manager Integrations tab showing two runtime entries: Generic CPU and Intel, with a tooltip explaining that the Intel library must be installed on the server"><figcaption><p>The Intel runtime appears in the list once the OpenVINO drivers are installed and the server has been restarted</p></figcaption></figure>

## Monitoring: on Ubuntu

Monitor the usage of Intel CPU, GPU, and NPU using the appropriate command:

| Hardware | Command              | How to install                            |
| -------- | -------------------- | ----------------------------------------- |
| CPU      | `top`                | Comes preinstalled in the OS              |
| GPU      | `sudo intel_gpu_top` | `sudo apt install intel-gpu-tools`        |
| NPU      | `intel-npu-top`      | <https://pypi.org/project/intel-npu-top/> |

## Notes: Incompatible Models

Not all models running on CPU are supported by the GPU and NPU; in that case, the AI Manager fails to run and displays an error message.\
For example, the face detection model works on CPU and GPU but not on NPU.\
Contact the [support team](/ai-manager-v6.1.4/support-and-troubleshooting/how-to-get-support.md) to check for a workaround in such situations.<br>