R
RunPod10mo ago
AC_pill

RTX 4090 POD Cuda issue

Hi, I'm trying to load a community POD and I run into this issue with the Cuda drives ERROR: The NVIDIA Driver is present, but CUDA failed to initialize. GPU functionality will not be available. 2024-02-08T16:56:04.457515868Z [[ Initialization error (error 3) ]] POD is US, RTX 4090 It's a NVidia container, details below: ============= 2024-02-08T16:56:04.419245167Z == PyTorch == 2024-02-08T16:56:04.419246369Z ============= 2024-02-08T16:56:04.419247661Z 2024-02-08T16:56:04.419248773Z NVIDIA Release 23.10 (build 71422337) 2024-02-08T16:56:04.419250697Z PyTorch Version 2.1.0a0+32f93b1 Is there any specific Cuda version for 4090s? The same Docker runs without issues on A5000, 3090.
6 Replies
Madiator2011
Madiator201110mo ago
Install PyTorch Nightly
ashleyk
ashleyk10mo ago
Or just use RunPod PyTorch template that installs a stable version of PyTorch
AC_pill
AC_pillOP10mo ago
thanks guys, will try both
kopyl
kopyl10mo ago
Did anything help you?
AC_pill
AC_pillOP10mo ago
Hi @kopyl yes, and what I did actually was updating my docker to use the latest 12.2. If you use the official Nvidia dockers, the latest NGC TensorRT, they are working just fine, thank you
kopyl
kopyl10mo ago
Cuda 11.8 was working with TensorRT for me too
Want results from more Discord servers?
Add your server