Error when compiling with CUDA on Windows about node-llama-cpp HOT 4 CLOSED

@BennyH26 There are many environment variables that you can set during the build and runtime to configure the CUDA support of llama.cpp; you can find more details here.

I do not have an Nvidia GPU, so I can't test how well the CUDA support works or provide better documentation for this process by myself.

I saw that someone else had trouble with the CUDA support and found a solution to their issue, I think it can help you: #22 (comment)

Please update if you managed to solve this issue, as I would like to document these problems and solutions to make it more straightforward to set up.

from node-llama-cpp.

Closed due to inactivity, as I assume this issue was fixed as part of #37, and I've got confirmation that CUDA works (on Linux though).

@BennyH26 Please update whether CUDA support now works for you on Windows.

from node-llama-cpp.

Closed due to inactivity, as I assume this issue was fixed as part of #37, and I've got confirmation that CUDA works (on Linux though).

@BennyH26 Please update whether CUDA support now works for you on Windows.

I have the exact same error trying to install this library with CUDA support.

from node-llama-cpp.

@AngryBacteria Do you still experience this issue with version 2.7.2?
If the issue still persists, please read the CUDA guide and let me know if it helped you or whether you still experience this issue.

from node-llama-cpp.