[Unit] Description=llama-server (aux/classification) โ€” Phi-4-14B Q4_K_M (OpenAI-compatible inference) After=network.target nvidia-persistenced.service Wants=nvidia-persistenced.service [Service] Type=simple User={{ llm_service_user }} Group={{ llm_service_user }} Environment="HOME=/home/{{ llm_service_user }}" ExecStart={{ llm_binary_path }} \ --model {{ llm_aux_model_path }} \ --host {{ llm_bind_address }} \ --port {{ llm_aux_port }} \ --ctx-size {{ llm_aux_ctx_size }} \ --n-gpu-layers {{ llm_aux_gpu_layers }} \ --parallel {{ llm_aux_parallel }} \ --metrics # NOTE: no --chat-template flag โ€” let llama-server auto-detect Phi-4's own # embedded chat template from GGUF metadata (same reasoning as the existing # llm-inference role's Gemma unit: explicit overrides risk mismatching the # model's actual expected format). # NOTE: --host is the private LAN IP (10.1.71.130 by default), NOT 0.0.0.0 โ€” # deliberate change from the pre-existing Gemma pattern (plan ยง5). Restart=on-failure RestartSec=10 TimeoutStartSec=600 StandardOutput=journal StandardError=journal SyslogIdentifier=llama-server-aux [Install] WantedBy=multi-user.target