No GPU is available for llama.cpp in sway

swaymsg -t get_version sway version 1.12

sway.log empty

Install https://github.com/ggml-org/llama.cpp

launch terminal emulator in sway

Issue a command:

llama-cli --list-devices Available devices: (none)

Expected result: GPU found llama-cli --list-devices Available devices: Vulkan0: Intel(R) Iris(R) Xe Graphics (TGL GT2) (11795 MiB, 10616 MiB free)

Actual result No GPU found

Note: no such issue in labwc