Concrete Compiler: GPU Enabled and Available Both False – How to Fix?

Zama Concrete ML
1

Hello,

I followed the tutorial on enabling GPU acceleration with Concrete ML and installed the necessary libraries. However, when I check, it returns False. Am I missing a library or doing something wrong?

Thank you in advance for your help!

2

Could you post the code which returns the False value ? Also please post the output of nvidia-smi and the version of cuda that you have installed - the output of find /usr -name "libcuda*"

3

The code return false:

import concrete.compiler;

print("GPU enabled: ", concrete.compiler.check_gpu_enabled())

print("GPU available: ", concrete.compiler.check_gpu_available())

nvidia - smi output:
Mon Mar 24 22:55:31 2025
±----------------------------------------------------------------------------------------+
| NVIDIA-SMI 570.124.04 Driver Version: 570.124.04 CUDA Version: 12.8 |
|-----------------------------------------±-----------------------±---------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA GeForce RTX 4060 … Off | 00000000:01:00.0 On | N/A |
| N/A 41C P8 5W / 50W | 615MiB / 8188MiB | 84% Default |
| | | N/A |
±----------------------------------------±-----------------------±---------------------+

±----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 2366 G Hyprland 203MiB |
| 0 N/A N/A 2502 G Xwayland 25MiB |
| 0 N/A N/A 3015 G …e8be37bc43f03ace8b2228d90641e 194MiB |
| 0 N/A N/A 5287 G kitty 35MiB |
| 0 N/A N/A 5761 G kitty 59MiB |
±----------------------------------------------------------------------------------------+

My cuda output:
/usr/lib64/libcuda.so
/usr/lib64/libcuda.so.1
/usr/lib64/libcuda.so.570.124.04
/usr/lib64/libcudadebugger.so.1
/usr/lib64/libcudadebugger.so.570.124.04