Elastic Container Instance:Create text-based light and shadow art by using a Stable Diffusion extension

Visit Hugging Face and obtain the IDs of the models.

In this example, the following models are used. After you find the models in Hugging Face, you can copy the model IDs at the top of the model details pages.

• Stable Diffusion model: hanafuusen2001/BeautyProMix

• ControlNet model: ioclab/ioc-controlnet

Log on to the Elastic Container Instance console.

In the top navigation bar, select a region.

In the navigation pane on the left, click Data Cache.

Create data caches for the models.

Click Create Data Cache to create data caches for the hanafuusen2001/BeautyProMix and ioclab/ioc-controlnet models separately. The following table describes the sample parameter values that are used in this example. You must use the specified values for the Cache Data Source parameter. These specified values are fixed configurations for pulling the models. You can specify other parameters as needed. For more information, see Create a data cache.

View the status of the data caches.

On the Data Cache page, refresh the page to view the status of the data caches. If the status becomes Available, the data caches are ready for use.

Copy the data caches.

If a data cache is in a different region from your application, you can find the target data cache on the Data Cache page and click Copy to copy it to the required region.

Visit Hugging Face and obtain the IDs of the models.

In this example, the following models are used. After you find the models in Hugging Face, you can copy the model IDs at the top of the model details pages.

• Stable Diffusion model: hanafuusen2001/BeautyProMix

• ControlNet model: ioclab/ioc-controlnet

Call the CreateDataCache API operation to create data caches.

1. Create a data cache for the hanafuusen2001/BeautyProMix model.

   The following code provides sample parameter values. The system pulls the model data from Hugging Face and saves it to the /model/beautypromix/ directory of the bucket named test. The data cache is named beautypromix and is retained for one day. When the system pulls the data from the model to create the data cache, it automatically creates an elastic IP address (EIP) and associates the EIP with the Virtual Private Cloud (VPC) that the data cache belongs to. This allows the system to pull data over the internet. After the data cache is created, the EIP is automatically released.

   {
     "RegionId": "cn-hongkong",
     "Path": "/model/beautypromix/",
     "Name": "beautypromix",
     "DataSource": {
       "Type": "URL",
       "Options": {
         "#10#repoSource": "HuggingFace/Model",
         "#6#repoId": "hanafuusen2001/BeautyProMix"
       }
     },
     "RetentionDays": 1,
     "EipCreateParam": {
       "Bandwidth": 5
     },
     "Bucket": "test"
   }

2. Create a data cache for the ioclab/ioc-controlnet model.

   The following code provides sample parameter values. The system pulls the model data from Hugging Face and saves it to the /model/ioc/ directory of the bucket named test. The data cache is named ioc and is retained for one day. When the system pulls the data from the model to create the data cache, it automatically creates an EIP and associates the EIP with the VPC that the data cache belongs to. This allows the system to pull data over the internet. After the data cache is created, the EIP is automatically released.

   {
     "RegionId": "cn-hongkong",
     "Path": "/model/ioc/",
     "Name": "ioc",
     "DataSource": {
       "Type": "URL",
       "Options": {
         "#10#repoSource": "HuggingFace/Model",
         "#6#repoId": "ioclab/ioc-controlnet"
       }
     },
     "RetentionDays": 1,
     "EipCreateParam": {
       "Bandwidth": 5
     },
     "Bucket": "test"
   }

Query the status of the data caches.

Use the returned DataCache ID to call the DescribeDataCaches API operation and query the information of the DataCache. If the status of the DataCache that is indicated by DataCaches.Status is Available, the DataCache is ready for use.

Copy the data caches.

If the data cache is in a different region from where you want to deploy the application, you can call the CopyDataCache API operation to copy the data cache to the required region.

On the Container Group page in the Elastic Container Instance console, click Create Container Group.

Configure the parameters that are used to create the container group (elastic container instance) and then click Confirm Configuration.

The following table describes the sample parameter values used to create the Elastic Container Instance. The instance is created based on a GPU-accelerated ECS instance type and is mounted with the data caches for the hanafuusen2001/BeautyProMix and ioclab/ioc-controlnet models. The container in the Elastic Container Instance uses an image provided by the Elastic Container Instance service that contains the Stable Diffusion application. After the container starts, it runs the python3 launch.py --listen --skip-torch-cuda-test --port 8888 --no-half command to start the WebUI.

Check the configurations of the instance, read and select the Terms of Service, and then click Confirm Order.

Return to the Container Group page, check whether the application is deployed, and view the EIP of the instance.

On the Container Group page, you can view the instance status. Click an instance ID to open the instance details page and view the container status. The application is successfully deployed when both the instance and container statuses are Running. You can find the EIP address of the instance in the IP Address column.

Use the data caches to create an elastic container instance. Deploy the Stable Diffusion application and the extension of the application on the instance.

The following example shows the parameters used to create an ECI instance by calling the CreateContainerGroup API operation. The instance uses a GPU specification and is mounted with the data caches for hanafuusen2001/BeautyProMix and ioclab/ioc-controlnet. The container in the instance uses an image that is provided by ECI and contains the Stable Diffusion application. After the container starts, it runs the python3 launch.py --listen --skip-torch-cuda-test --port 8888 --no-half command to start the WebUI.

{
  "RegionId": "cn-beijing",
  "SecurityGroupId": "sg-2ze7l1o0ql1cbk******",
  "VSwitchId": "vsw-2ze23nqzig8inpr******",
  "ContainerGroupName": "sd-web",
  "InstanceType": "ecs.gn6i-c16g1.4xlarge",
  "DataCacheBucket": "test",
  "Container": [
    {
      "Arg": [
        "-c",
        "python3 launch.py --listen --skip-torch-cuda-test --port 8888 --no-half"
      ],
      "Command": [
        "/bin/sh"
      ],
      "Gpu": 1,
      "Name": "sd",
      "Image": "registry.cn-hangzhou.aliyuncs.com/eci_open/stable-diffusion:1.0.1",
      "VolumeMount": [
        {
          "Name": "model1",
          "MountPath": "/stable-diffusion-webui/models/Stable-diffusion/"
        },
        {
          "MountPath": "/stable-diffusion-webui/extensions/sd-webui-controlnet/models/",
          "Name": "model2"
        }
      ]
    }
  ],
  "Volume": [
    {
      "Type": "HostPathVolume",
      "HostPathVolume.Path": "/model/beautypromix/",
      "Name": "model1"
    },
    {
      "Type": "HostPathVolume",
      "HostPathVolume.Path": "/model/ioc/",
      "Name": "model2"
    }
  ],
  "DataCacheProvisionedIops": 35000,
  "DataCacheBurstingEnabled": true,
  "AutoCreateEip": true
}

Check whether the application is deployed.

Use the returned instance ID to call the DescribeContainerGroupStatus API operation and query the status of the instance and the container. If the status of the instance that is indicated by Status and the status of the container that is indicated by ContainerStatuses.State are Running, the instance is created and the container is running.

Query the EIP of the elastic container instance.

Use the returned instance ID to call the DescribeContainerGroups API operation and query the instance details. You can obtain the EIP of the instance from the InternetIP parameter.