[Unit]
Description=A Llama CPP Server Running GPT OSS 120b

[Container]
Pod=ai.pod
Image=localhost/llama-cpp-vulkan:2026-01-12-10-13-30
Volume=llama-server-cache:/root/.cache
AddDevice=/dev/kfd
AddDevice=/dev/dri

Exec=-hf ggml-org/gpt-oss-120b-GGUF \
--ctx-size 32000 \
--jinja \
-ub 2048 \
-b 2048 \
--port 8000 \
--host 0.0.0.0 \
-n -1 \
--n-gpu-layers 999

[Service]
Restart=always
# Extend Timeout to allow time to pull the image
TimeoutStartSec=900

[Install]
# Start by default on boot
WantedBy=multi-user.target default.target