1：修改 复制  ./dify/docker/.env.example  ./dify/docker/.env 添加一下内容

# 启用自定义模型
CUSTOM_MODEL_ENABLED=true

# 将OLLAMA_API_BASE_URL 改为宿主机的物理ip
OLLAMA_API_BASE_URL=http://192.168.72.8:11434

# vllm 的 OPENAI的兼容 API 地址
CUSTOM_MODEL_API_BASE_URL=http://192.168.72.8:8000

# 其中192.168.72.8 为宿主机的物理IP，端口根据自己的实际情况确定

2：修改 dify/docker/docker-compose.yaml

修改serivces 下的内容

services:
  # API service
  api:
    image: langgenius/dify-api:0.15.3
    restart: always
    environment:
      # Use the shared environment variables.
      <<: *shared-api-worker-env
      # Startup mode, 'api' starts the API server.
      MODE: api
      SENTRY_DSN: ${API_SENTRY_DSN:-}
      SENTRY_TRACES_SAMPLE_RATE: ${API_SENTRY_TRACES_SAMPLE_RATE:-1.0}
      SENTRY_PROFILES_SAMPLE_RATE: ${API_SENTRY_PROFILES_SAMPLE_RATE:-1.0}
    depends_on:
      - db
      - redis
    volumes:
      # Mount the storage directory to the container, for storing user files.
      - ./volumes/app/storage:/app/api/storage
    networks:
      - ssrf_proxy_network
      - default
      - dify-net
    extra_hosts:
      - "host.docker.internal:192.168.72.8"  # 宿主机地址

修改worker 下的内容

worker:
    image: langgenius/dify-api:0.15.3
    restart: always
    environment:
      # Use the shared environment variables.
      <<: *shared-api-worker-env
      # Startup mode, 'worker' starts the Celery worker for processing the queue.
      MODE: worker
      SENTRY_DSN: ${API_SENTRY_DSN:-}
      SENTRY_TRACES_SAMPLE_RATE: ${API_SENTRY_TRACES_SAMPLE_RATE:-1.0}
      SENTRY_PROFILES_SAMPLE_RATE: ${API_SENTRY_PROFILES_SAMPLE_RATE:-1.0}
    depends_on:
      - db
      - redis
    volumes:
      # Mount the storage directory to the container, for storing user files.
      - ./volumes/app/storage:/app/api/storage
    networks:
      - ssrf_proxy_network
      - default
      - dify-net
    extra_hosts:
      - "host.docker.internal:192.168.72.8"  # 宿主机地址

修改networks 下的内容

networks:
  # create a network between sandbox, api and ssrf_proxy, and can not access outside.
  ssrf_proxy_network:
    driver: bridge
    internal: true
  milvus:
    driver: bridge
  opensearch-net:
    driver: bridge
    internal: true
  dify-net:
    driver: bridge
    internal: true

重启docker 容器

sudo docker-compose down
docker compose up -d

二： dify 上添加模型

（1）右上角用户名，点解设置

（2）选择模型供应商，找到相应的模型供应商

（3）填入模型类型、模型名称、url 等信息后点击保存

（4）确认模型添加成功

添加成功后在模型列表中可以看到已经添加的模型

创建应用进行效果测试

五 添加知识库

数据分块查看

六 应用绑定数据库