Search results for query: nvidia-smi

  1. B

    vGPU doesn't work with pytorch/nccl/vllm

    sure! $ nvidia-smi Fri Nov 7 14:13:38 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 580.95.05 Driver Version: 580.95.05 CUDA Version: 13.0 |...
  2. W

    vGPU doesn't work with pytorch/nccl/vllm

    Did you chose correct vgpu profile and set Nvidia license token in vm? Do you mind to show nvidia-smi nvidia-smi -q From vm?
  3. T

    Passthrough 4090 Lockup Linux Only

    ...One thing I did notice is that, after installing the nvidia driver on the host machine, if I tried to unbind from vfio and bind to nvidia, nvidia-smi wouldn't show the 4090, so I did lspci -nnk -d 10de: and it showed the active driver being nvidia for both the 4090 and the 970. But nvidia-smi...
  4. B

    vGPU doesn't work with pytorch/nccl/vllm

    ...for VDIs. For LLM Inference we have mapped 4 vGPUs into a Virtual machine. All 4 vGPUs show up correctly in the guest system (using nvidia-smi). We setup vLLM (0.11.0) and torch 2.8.0+cu128 in a python virtual environment. To our understanding torch comes with a pre-compiled cuda + nccl...
  5. K

    Power Consumption when GPU idle with Passthrough

    ...would stay in Power State P0 - turns out the default persistenced mode for the daemon on my install was "off" (you can verify by running "nvidia-smi -pm 1" to turn it on and see the difference). I modified /etc/systemd/system/nvidia-persistenced.service to set the default as...
  6. T

    [SOLVED] NVIDIA vGPU - No devices were found

    ...erkannt. ./NVIDIA-Linux-x86_64-580.95.02-vgpu-kvm.run --accept-license --no-questions --ui=none --kernel-module-type=proprietary --dkms nvidia-smi Mon Nov 3 09:48:41 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 580.95.02...
  7. T

    [SOLVED] NVIDIA vGPU - No devices were found

    ...mit NVIDIA Grid. Auf einem SuperMicro GPU Host wird nach Update der Grid Host Treiber von 550 auf 570 unsere RTX A5000 nicht mehr über nvidia-smi erkannt. Nach Downgrade auf 550 funktioniert wieder alles einwandfrei. Proxmox 8.4 (letzter 8.X Stand) Kernel: 6.8.12-15-pve Es sind mehrere...
  8. D

    3 Minute Delay Starting VM with GPU Passthrough (vfio-pci reset issue)

    ...|| true sleep 1 # Kurze Pause # GPU auf dem Host initialisieren (genau wie in Ihrem manuellen Script) nvidia-smi || true log "GPU erfolgreich an Host zurückgegeben." ;; esac exit 0 There is no vendor-reset like for AMD GPUs for NVIDIA is it?
  9. D

    3 Minute Delay Starting VM with GPU Passthrough (vfio-pci reset issue)

    ...sleep 2 echo "$GPU" > /sys/bus/pci/drivers/nvidia/bind echo "$AUDIO" > /sys/bus/pci/drivers/snd_hda_intel/bind nvidia-smi >/dev/null 2>&1 log "GPU successfully returned to host." ;; esac exit 0 When I shut down the VM, the post-stop script fails with "Device or resource...
  10. J

    Opt-in Linux 6.14 Kernel for Proxmox VE 8 available on test & no-subscription

    If you intend to use the gpu on the host or in LXC containers, as opposed to passing through to a VM, don’t blacklist nvidia*
  11. R

    Opt-in Linux 6.17 Kernel for Proxmox VE 9 available on test & no-subscription

    ...Adding boot menu entry for UEFI Firmware Settings ... done Setting up proxmox-kernel-6.17 (6.17.1-1) ... root@pve-bdr:~# nvidia-smi Wed Oct 15 07:12:54 2025 +-----------------------------------------------------------------------------------------+ | NVIDIA-SMI 580.95.05...
  12. P

    Opt-in Linux 6.14 Kernel for Proxmox VE 8 available on test & no-subscription

    ...updated to kernel 6.17, updated NVIDIA drivers from 570.x series to latest 580.x and I don't have any hardware transcoding, even though nvidia-smi shows GPU information on both host and in Plex container...
  13. D

    PVE 9 - NVIDIA Container Toolkit Broken

    Also, make sure you can run nvidia-smi in the container.
  14. D

    PVE 9 - NVIDIA Container Toolkit Broken

    Can you post the config of the LXC?
  15. Z

    PVE 9 - NVIDIA Container Toolkit Broken

    @dasunsrule32 I tried your method on a new unpriviledged lxc (debian 13). I deployed it with the docker lxc helper script. I can get nvidia-smi to work inside the container. However, I'm having problems with docker. Trying to run ollama, and getting this error: I've checked the devices inside...
  16. S

    [SOLVED] High Power Usage for Passthrough GPU When VM Is Stopped

    This is a great write up pretty much what I did. just a note the NVIDIA driver comes with the persistence scrip you just need to install it you can find it here /usr/share/doc/NVIDIA_GLX-1.0/samples/nvidia-persistenced-init.tar.bz2
  17. Impact

    Finding GPU Passthrough LXC Path

    Please check that the GPU driver is loaded via lspci -vnnk | awk '/VGA/{print $0}' RS= | grep -Pi --color "^|(?<=Kernel driver in use: |Kernel modules: )[^ ]+" Also check what nvidia-smi says. What does pct start CTIDHERE --debug say?
  18. N

    Install GPU on VM

    I setup and followed instruction on google also here but all is failed. My server is Gigabyte R282-z93, Dual EPYC 7642, RAM 256GB and GPU RTX5000 ADA. Proxmox 9 (initrd=\EFI\proxmox\6.14.11-2-pve\initrd.img-6.14.11-2-pve root=ZFS=rpool/ROOT/pve-1 boot=zfs amd_iommu=on iommu=pt video=efifb:off...
  19. C

    [SOLVED] High Power Usage for Passthrough GPU When VM Is Stopped

    ...echo "" > /sys/bus/pci/devices/0000:02:00.1/driver_override - After installation, both GPUs should be visible to the host again under ```nvidia-smi``` --- ## **Step 2 — Install NVIDIA Driver** 1. Make installer executable (this can be downloaded from Nvidia page): chmod +x...
  20. C

    Power Consumption when GPU idle with Passthrough

    ...can see it: For my gtx 1060 - For my Quadro P400 - After installation, both GPUs should be visible to the host again under ```nvidia-smi``` --- ## **Step 2 — Install NVIDIA Driver** 1. Make installer executable (this can be downloaded from Nvidia page): 2. Run installer: -...