version: '3.8'

services:
  # The Intelligence Router
  router:
    build: ./intelligence-router
    ports:
      - "9000:9000"
    environment:
      - MAIN_PC_URL=http://10.0.4.x:8080/v1
      - LOCAL_SLM_URL=http://llama-slm:8080/v1
      - OPENAI_API_KEY=${OPENAI_API_KEY}
    depends_on:
      - llama-slm

  # The Local SLM (Fallback Brain)
  llama-slm:
    image: ghcr.io/ggerganov/llama.cpp:server
    volumes:
      - ./models:/models
    # Command to run a small, fast model (e.g., Llama-3-8B GGUF)
    command: >
      -m /models/llama-3-8b-instruct.Q4_K_M.gguf
      --host 0.0.0.0
      --port 8080
      --ctx-size 2048
    ports:
      - "8081:8080"