# VM 5702 — Inference: llama.cpp server (CPU-friendly)
# Copy to /opt/ai/inference/ and place model at /opt/ai/inference/data/models/model.gguf
# See: docs/02-architecture/AI_AGENTS_57XX_DEPLOYMENT_PLAN.md Appendix D

services:
  llama:
    image: ghcr.io/ggerganov/llama.cpp:server
    container_name: ai-inf-prod
    volumes:
      - /opt/ai/inference/data/models:/models
    command: >
      -m /models/model.gguf
      --host 0.0.0.0 --port 8000
      --n-gpu-layers 0
      --ctx-size 4096
    ports:
      - "8000:8000"
    restart: unless-stopped