Tensorflow with NVIDIA Corporation GK107GL [GRID K1]

Hi, i want to do tensorflow inference with NVIDIA Corporation GK107GL [GRID K1] on Ubuntu 18.04
it`s possible? I tryed to install:

  1. nvidia-340
  2. cuda-6.5
  3. Tensorflow-1.0.5 but tensorflow 1.0.5 i saw it`s require cuda 8 at least.
  4. After install cuda i lose desktop-gui

ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:05.0 ==
modalias : pci:v000010DEd00000FF2sv000010DEsd00001012bc03sc00i00
vendor : NVIDIA Corporation
model : GK107GL [GRID K1]
driver : nvidia-340 - distro non-free recommended
driver : nvidia-304 - third-party non-free
driver : xserver-xorg-video-nouveau - distro free builtin

VM-ubuntu18:~$lspci -vnnn | perl -lne ‘print if /^\d+:.+([\S+:\S+])/’ | grep VGA

00:01.0 VGA compatible controller [0300]: Cirrus Logic GD 5446 [1013:00b8] (prog-if 00 [VGA controller])
00:05.0 VGA compatible controller [0300]: NVIDIA Corporation GK107GL [GRID K1] [10de:0ff2] (rev a1) (prog-if 00 [VGA controller])

VM-ubuntu18:~$ nvidia-smi
Fri Aug 13 08:49:49 2021
±-----------------------------------------------------+
| NVIDIA-SMI 340.108 Driver Version: 340.108 |
|-------------------------------±---------------------±---------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GRID K1 Off | 0000:00:05.0 Off | N/A |
| N/A 35C P0 12W / 31W | 10MiB / 4095MiB | 0% Default |
±------------------------------±---------------------±---------------------+

±----------------------------------------------------------------------------+
| Compute processes: GPU Memory |
| GPU PID Process name Usage |
|=============================================================================|
| No running compute processes found |
±----------------------------------------------------------------------------+

Hi @mihaic
Please note this forum branch is dedicated to CUDA GDB support. You question might be more suitable for (for example): CUDA Setup and Installation - NVIDIA Developer Forums