Parameters passed to Alibaba Cloud Model Studio - Intelligent Media Services

This topic describes the parameters passed from AI real-time interaction to Alibaba Cloud Model Studio. It helps developers configure and pass parameters for agent orchestration applications.

Parameters

Parameter	Type	Description	Example
biz_params	Object	The custom parameters of a workflow application and agent orchestration application. To pass parameters to a plug-in, add `user_defined_params`. For more information, see Parameter pass-through for applications. Note Alibaba Cloud passes the following agent-related parameters. You do not need to configure them. instanceId: the ID of the agent instance. channelId: the ID of the RTC channel.	To a plug-in `{ "user_defined_params": { "city":"Singapore" } }` To a workflow `{ "city":"Singapore" }`
memory_id	String	A unique identifier for long-term memory. This parameter enables efficient preservation and recall of historical information.	YOUR_MEMORY_ID
image_list	Array	The URLs to associated images that provide visual information related to the current operation.	["https://example.com/images/example.jpg"]
rag_options	Object	Parameters related to Retrieval-Augmented Generation (RAG). This includes but is not limited to retrieving specified knowledge bases or documents.
pipeline_ids	Array	The knowledge base ID. When specified, all documents within the specified knowledge base will be retrieved.	["Knowledge base ID1", "Knowledge base ID2"]
file_ids	Array	The ID of the unstructured document to be processed.	["Document ID1", "Document ID2"]
metadata_filter	Object	The metadata of unstructured documents. When specified, documents that contain the metadata are retrieved.	{ "name": "Zhang San" }
structured_filter	Object	The column names and the corresponding values that are used to filter structured documents.	{ "key1": "value1", "key2": "value2" }
tags	Array	The tags associated with unstructured documents. When specified, unstructured documents that contain the tags are retrieved.	["Tag 1", "Tag 2"]

For more information, see Use API to call application.

Parameter examples

Plug-in

bailianParamsJson = {
    "biz_params": {
        "user_defined_params": 
        {
            "your_plugin_id":
            {
                "article_index": 2
            }
        }
    },
    "memory_id": "your_memory_id",
    "image_list": [
        "https://your_image_url"  # URL to an image
    ],
    "rag_options": {
        "pipeline_ids": [
            "your_id",  # Knowledge base ID
        ],
        "file_ids": [
            "Document ID1",  # Document ID
            "Document ID2"   # Document ID
        ],
        "metadata_filter": {
            "name": "Zhang San"  # Filter condition of metadata
        },
        "structured_filter": {
            "key1": "value1",  # Filter condition of structured documents
            "key2": "value2"   # Filter condition of structured documents
        },
        "tags": [
            "Tag 1",  # Tag
            "Tag 2"   # Tag
        ]
    }
}
bailianParams = json.dumps(bailianParamsJson)

Workflow


bailianParamsJson = {
    "biz_params": {
        "key1": "value1"
    },
    "memory_id": "your_memory_id",
    "image_list": [
        "https://your_image_url"  # URL to an image
    ],
    "rag_options": {
        "pipeline_ids": [
            "your_id"  # Knowledge base ID
        ],
        "file_ids": [
            "Document ID1",  # Document ID
            "Document ID2"  # Document ID
        ],
        "metadata_filter": {
            "name": "Zhang San"  # Filter condition of metadata
        },
        "structured_filter": {
            "key1": "value1",  # Filter condition of structured documents
            "key2": "value2"  # Filter condition of structured documents
        },
        "tags": [
            "Tag 1",  # Tag
            "Tag 2"  # Tag
        ]
    }
}
bailianParams = json.dumps(bailianParamsJson)