I’ve downloaded and installed NVIDIA driver for it.
nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2021 NVIDIA Corporation
Built on Fri_Dec_17_18:28:54_Pacific_Standard_Time_2021
Cuda compilation tools, release 11.6, V11.6.55
Build cuda_11.6.r11.6/compiler.30794723_0
nvidia-smi
Wed Jun 12 19:45:08 2024
±----------------------------------------------------------------------------------------+
| NVIDIA-SMI 555.85 Driver Version: 555.85 CUDA Version: 12.5 |
|-----------------------------------------±-----------------------±---------------------+
| GPU Name Driver-Model | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 Tesla M60 TCC | 00000000:00:1D.0 Off | 0 |
| N/A 33C P0 41W / 150W | 7372MiB / 7680MiB | 0% Default |
| | | N/A |
±----------------------------------------±-----------------------±---------------------+
| 1 Tesla M60 TCC | 00000000:00:1E.0 Off | 0 |
| N/A 28C P8 14W / 150W | 10MiB / 7680MiB | 0% Default |
| | | N/A |
±----------------------------------------±-----------------------±---------------------+
±----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 3724 C …Programs\Python\Python39\python.exe 7360MiB |
±----------------------------------------------------------------------------------------+
torch (1.12.0+cu116) is able to detect the 2 GPUs.
Num GPUs Available: 2
CUDA is available: True
CUDA version: 11.6
However, it does not want to “play” with transformers (4.42.0.dev0) /tensorflow (2.5.0), it keeps on complaining
“RuntimeError: CUDA error: out of memory
CUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.”
And btw, CUDNN, 9.2 has been installed as well. And “CUDA_HOME”, additions to system path etc. have all been setup.
How to get this beast to work?
Thanks.