-
Notifications
You must be signed in to change notification settings - Fork 11.4k
OpenCL can select discrete GPU by default? #1546
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
It is possible to select the OpenCL device with an environment variable: export GGML_OPENCL_DEVICE=1
./main .... You can use numbers 0, 1, 2 ... or you can use string matching, like |
Actually, I need to write the documentation for this part. It doesn't seem to be anywhere. |
Is there a way to do this for cuda devices? I have no idea how to tell it to use my other gpu |
You just list the devices that are visible to the application: CUDA_VISIBLE_DEVICES=1 ./main ... |
README now should have info about GPU selection for CUDA and OpenCL. |
I have Old MacBook Pro with one intel GPU and one AMD discrete GPU. I am using OpenCL ggml, and ggml default choose Intel GPU. I hope ggml can using discrete GPU by default, or we can set GPU device though API?
The text was updated successfully, but these errors were encountered: