# Mixing A100 and RTX3090 in the same server

**URL:** <https://forums.developer.nvidia.com/t/mixing-a100-and-rtx3090-in-the-same-server/235282>\
**Category:** Drivers - Linux, Windows, MacOS\
**Tags:** hw, cuda\
**Created:** [November 24, 2022, 7:24am UTC](https://forums.developer.nvidia.com/t/mixing-a100-and-rtx3090-in-the-same-server/235282 "2022-11-24T07:24:31Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![ryuu.j.ching](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@ryuu.j.ching](https://forums.developer.nvidia.com/u/ryuu.j.ching)\
**Post date:** [November 24, 2022, 7:24am UTC](https://forums.developer.nvidia.com/t/mixing-a100-and-rtx3090-in-the-same-server/235282/1 "2022-11-24T07:24:31Z")

</div>

Hi, all

Currently, I have a couple of 32GB V100 in _server A_, and couple of A100 in _server B_, both servers have a few extra empty PCIe lanes. I am considering adding some of RTX 3090 or 4090 cards to fill the empty lanes.

I’m wondering if these combinations (V100+RTX, A100+RTX) will cause some issue when doing some deep-learning stuff. Do these cards support the same Nvidia driver and CUDA driver?

Thanks
