> For the complete documentation index, see [llms.txt](https://docs.cloudeka.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cloudeka.ai/deka-gpu/deka-gpu-vms/gpu-type.md).

# GPU Type

{% hint style="warning" %}
You cannot install the **NVIDIA L40S Q-Series Driver** independently. You need to use the **Ticket** feature on the **Service Portal Cloudeka** to ask our team for help with installing NVIDIA drivers and activating NVIDIA licenses. For more information on how to create tickets to request assistance from the Cloudeka team, click [this link](https://docs.cloudeka.ai/deka-gpu-ticket/create-ticket).
{% endhint %}

The NVIDIA L40S offers outstanding multi-workload performance with the Ada Lovelace architecture, combining powerful AI computing with the best graphics and media acceleration. With its fourth-generation Tensor Cores and 48 GB of memory, the L40S is ideal for AI training and inference, 3D rendering, and video workloads.

There are several types of GPUs that can be used on the **Service Portal Cloudeka**.

<table><thead><tr><th width="213">Type</th><th width="204">GPU Memory</th><th>Code Name</th></tr></thead><tbody><tr><td>L40s Passthrough</td><td>48 GB</td><td>nvidia.com/AD102GL_L40S</td></tr><tr><td>L40s 48Q</td><td>48 GB</td><td>nvidia.com/NVIDIA_L40S-48Q</td></tr><tr><td>L40s 24Q</td><td>24 GB</td><td>nvidia.com/NVIDIA_L40S-24Q</td></tr><tr><td>L40s 16Q</td><td>16 GB</td><td>nvidia.com/NVIDIA_L40S-16Q</td></tr><tr><td>L40s 8Q</td><td>8 GB</td><td>nvidia.com/NVIDIA_L40S-8Q</td></tr></tbody></table>