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

[Container]
# Image is built locally via podman build
Image=localhost/llama-cpp-vulkan:latest

# Downloaded models volume
Volume=/home/ai/models:/models:z

# Ports
PublishPort=8010:8010

# GPU Device
AddDevice=/dev/kfd
AddDevice=/dev/dri

# Server command
Exec=--port 8010 \
    -m /models/embedding/nomic-embed-text-v2/nomic-embed-text-v2-moe-q8_0.gguf \
    -ngl all \
    -ub 2048 \
    -b 2048 \
    --ctx-size 2048 \
    --embeddings \
    --models-dir /models

[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