CUDA error: no kernel image is available for execution on the device #196
Unanswered
NigelNelson
asked this question in
Q&A
Replies: 1 comment 2 replies
-
I could suggest setting |
Beta Was this translation helpful? Give feedback.
2 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I am attempting to use Exllama on a unique device. My platform is aarch64 and I have a NVIDIA A6000 dGPU. As such, the only compatible torch 2.1.0 build I can find is one for Python 3.8. The only issues that arose from this were Python 3.10 type checks failing, which I fixed by importing the relevant types from
typing
. My issue is when the torch extension is built, all I see in~/.cache/torch_extensions/
ispy38_cu114
notpy38_cu118
which is what I was expecting. Does anyone know how these extensions are built so I can potentially hack it for my use case? Torch.cuda.is_available() returns True.python==3.8.17
torch==2.1.0a0+41361538.nv23.06
cuda==11.8
Stacktrace:
Beta Was this translation helpful? Give feedback.
All reactions