# CUDA driver version is insufficient for CUDA runtime version

**URL:** <https://forums.developer.nvidia.com/t/cuda-driver-version-is-insufficient-for-cuda-runtime-version/55062>\
**Category:** CUDA Setup and Installation\
**Created:** [November 10, 2017, 7:23am UTC](https://forums.developer.nvidia.com/t/cuda-driver-version-is-insufficient-for-cuda-runtime-version/55062 "2017-11-10T07:23:39Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jinlingge](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@jinlingge](https://forums.developer.nvidia.com/u/jinlingge)\
**Post date:** [November 10, 2017, 7:23am UTC](https://forums.developer.nvidia.com/t/cuda-driver-version-is-insufficient-for-cuda-runtime-version/55062/1 "2017-11-10T07:23:39Z")

</div>

Dear all,

I accidentally purged my nvidia driver (375.82), a big mistake…  
Then I tried to reinstall it using: sudo apt-get install nvidia-375

After the installation, running ./deviceQuery from cuda samples, it pops out：

$ ./deviceQuery  
./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

cudaGetDeviceCount returned 35  
 → CUDA driver version is insufficient for CUDA runtime version  
Result = FAIL

The info from running nvidia-smi are:

±----------------------------------------------------------------------------+  
| NVIDIA-SMI 375.82 Driver Version: 375.82 |  
|-------------------------------±---------------------±---------------------+  
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |  
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |  
|===============================+======================+======================|  
| 0 GeForce GTX 105… Off | 0000:01:00.0 Off | N/A |  
| N/A 48C P0 N/A / N/A | 364MiB / 4038MiB | 13% Default |  
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+  
| Processes: GPU Memory |  
| GPU PID Type Process name Usage |  
|=============================================================================|  
| 0 1761 G /usr/lib/xorg/Xorg 234MiB |  
| 0 3628 G compiz 127MiB |  
±----------------------------------------------------------------------------+

The I tried to reinstall cuda 8.0 (./cuda\_8.0.44\_linux.run, downloaded from nvidia official website, ignoring the installation of cuda), the same problem still remains.

Some further information about my ubuntu environment are:

uname -a :  
Linux yj253pc 4.10.0-33-generic #37~16.04.1-Ubuntu SMP Fri Aug 11 14:07:24 UTC 2017 x86\_64 x86\_64 x86\_64 GNU/Linux

cat /proc/driver/nvidia/version :

NVRM version: NVIDIA UNIX x86\_64 Kernel Module 375.82 Wed Jul 19 21:16:49 PDT 2017  
GCC version: gcc version 4.9.3 (Ubuntu 4.9.3-13ubuntu2)

lspci | grep -i nvidia :

01:00.0 3D controller: NVIDIA Corporation Device 1c8c (rev a1)

Does anyone have any idea how to fix this problem?

Thanks a lot!

Best regards,

---

<div class="post-metadata">

**Author:** ![b3ta5sigma6rh0](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@b3ta5sigma6rh0](https://forums.developer.nvidia.com/u/b3ta5sigma6rh0)\
**Post date:** [January 25, 2018, 9:32am UTC](https://forums.developer.nvidia.com/t/cuda-driver-version-is-insufficient-for-cuda-runtime-version/55062/2 "2018-01-25T09:32:09Z")

</div>

This is much like the problem I am facing.  
Have you had any luck in solving it?