docker-compose.yml

version: '3.8'

services:
  text-gen:
    image: ghcr.io/huggingface/text-generation-inference:1.3
    ports:
      - "8080:80"
    environment:
      - MODEL_ID=HuggingFaceH4/zephyr-7b-beta
      - HF_HOME=/cache
      - HUGGINGFACE_HUB_CACHE=/cache
    volumes:
      - ./cache:/cache
    deploy:
      resources:
        limits:
          memory: 16G  # Ajusta si sabes cuánto puedes usar