EMR Serverless Spark includes several built-in models that you can use directly with an AI Function. If the built-in models do not meet your business needs, you can register an external model service. This lets you integrate with Model Studio, PAI-EAS, or your own self-managed model services.
Built-in models
EMR Serverless Spark includes the following built-in models. You can use them directly with an AI Function without registration.
Model service name | Model name |
qwen3.6-plus | qwen3.6-plus |
qwen3.5-plus | qwen3.5-plus |
qwen-plus | qwen-plus |
text-embedding-v4 | text-embedding-v4 |
tongyi-embedding-vision-plus | tongyi-embedding-vision-plus |
Add external models
If the built-in models do not suit your needs, follow these steps to add a self-managed model service.
Navigate to the model service page.
Log on to the E-MapReduce console.
In the left-side navigation pane, choose .
On the Spark page, click the name of your target workspace.
On the EMR Serverless Spark page, click in the left-side navigation pane.
On the Model Service tab, click Create External Model Service and configure the following parameters:
Parameter
Example value
Description
Model service name
my_qwen_serviceThis name is used for the
endpointNameparameter in anAI Function. It must be unique within the workspace and cannot be changed after creation.Endpoint
http://12*******39.vpc.cn-hangzhou.pai-eas.aliyuncs.com/api/predict/<ServiceName>/v1Enter the endpoint of the external model service. If you are using a PAI-EAS model, append /v1 to the end of the URL.
NoteIf the model service uses a public network endpoint, configure a Network Connection with public network access for the Spark job. For more information, see Network Connection.
Model name
Qwen3.5-PlusThe model name used in API calls.
Model type
ChatSelect
ChatorEmbeddingbased on the model you deploy.API key
nMzI**********************Zg==The API key for the model service.
Description
The latest Qwen multimodal model service
Enter a brief description for easy identification.
Verify your parameters and click create to register the model service.