# Quadro K2200 + Tesla K40C + Ubuntu **URL:** \ **Category:** CUDA Setup and Installation\ **Created:** [September 28, 2017, 10:08pm UTC](https://forums.developer.nvidia.com/t/quadro-k2200-tesla-k40c-ubuntu/53820 "2017-09-28T22:08:31Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![notdeveloper](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@notdeveloper](https://forums.developer.nvidia.com/u/notdeveloper)\ **Post date:** [September 28, 2017, 10:08pm UTC](https://forums.developer.nvidia.com/t/quadro-k2200-tesla-k40c-ubuntu/53820/1 "2017-09-28T22:08:31Z")
Please, help me! I’ve got experience in configuring systems, based on single GPU card. But I need to set up system with hardware: 1 X Quadro K2200 2 X Tesla K40C When I use search driver page, there is no appropriate driver for both. I need work with CUDA (i prefer version 8) + Caffe + other frameworks. Questions: 1. What driver I have to install for working both Tesla and Quadro? 2. Maybe, somewhere there is instruction/tutorial for setting up systems with multiple GPUs? 3. Which way is the best: - install drivers from CUDA Toolkit - install drivers from deb/run/repo? ---
**Author:** ![Robert\_Crovella](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/robert_crovella/32/14043_2.png) [@Robert\_Crovella](https://forums.developer.nvidia.com/u/Robert_Crovella)\ **Post date:** [September 28, 2017, 10:40pm UTC](https://forums.developer.nvidia.com/t/quadro-k2200-tesla-k40c-ubuntu/53820/2 "2017-09-28T22:40:28Z")
I would just install the CUDA 8 cuda toolkit. Whether you use the package manager method or the runfile installer method: [url][Installation Guide Linux :: CUDA Toolkit Documentation](http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html%5B/url%5D) you will get a driver that is compatible with both K2200 and K40c Whether you use runfile or package manager, get your installers from [url][http://www.nvidia.com/getcuda[/url]](http://www.nvidia.com/getcuda%5B/url%5D) now that CUDA 9 is released, CUDA 8 is no longer current. If you want CUDA 8, scroll down to the bottom of the page and use the “Legacy Releases” link.