全部產品
Search

搜尋本產品

    Vector Retrieval Service for Milvus：摘要

    文件中心

    Vector Retrieval Service for Milvus：摘要

    更新時間：Aug 04, 2026

    AI_SUMMARIZE 函數用於產生文本、映像或視頻的濃縮摘要，提升資訊擷取效率。整合於 Collection 時，可實現“寫入即摘要”，自動將源內容處理後存入目標文字欄位，無需額外後處理步驟。

    命令格式

    REST 介面

    POST /v2/vectordb/ai/summarize
Content-Type: application/json

{
  "model_name": "<模型名>",
  "texts": ["<文本或媒體URL>"],
  "params": {"max_words": 50, "media_type": "image"}
}

    Python

    schema = MilvusClient.create_schema(auto_id=True, enable_dynamic_field=False)
schema.add_field("id", DataType.INT64, is_primary=True)
schema.add_field("content", DataType.VARCHAR, max_length=8192)
schema.add_field("summary", DataType.VARCHAR, max_length=1024)
schema.add_field("dummy_vector", DataType.FLOAT_VECTOR, dim=2)
schema.add_function(
    Function(
        name="summarize_content",
        function_type=texttransform_function_type(),
        input_field_names=["content"],
        output_field_names=["summary"],
        params={
            "provider": "aliyun_milvus",
            "model_name": "<模型名>",
            "task": "ai_summarize",
            "max_words": "30",
            "temperature": "0",
        },
    )
)

    參數說明

    參數

    說明

    model_name

    必填。文本可使用已配置文本模型（如 qwen3.7-max）；圖片/視頻須使用已配置多模態模型（如 qwen3.7-plus）。

    texts

    REST 必填。待摘要文本，或圖片/視頻 URL。

    max_words

    選填。摘要最大詞數，預設 50;0 表示不限制。

    prompt

    選填。補充摘要要求，最大 5000 字元，不支援 ${...}。

    media_type

    選填。取值 image 或 video，表示輸入為媒體 URL。

    enable_thinking

    選填但**文本摘要建議設為 false**：文本模型（如 qwen3.7-max）開啟思考模式時，摘要任務易超過網關逾時而返回 request timeout(code 10001)。

    temperature / max_concurrency / timeout_sec

    選填。模型穩定性、並發和逾時設定。

    provider / task

    僅 Collection Function 必填，固定為 aliyun_milvus 與 ai_summarize。

    傳回值說明

    data.output.outputs 返回與輸入同序的摘要文本。Schema 寫入時，摘要寫入目標文字欄位。

    樣本一：為知識庫文章產生導讀

    知識庫需要在文章寫入時產生不超過 30 詞的首頁導讀。文本摘要使用文本模型 qwen3.7-max，並關閉思考模式以避免逾時。

    REST 介面

    #!/usr/bin/env bash
set -euo pipefail

MILVUS_REST_BASE_URL="http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_AUTH_TOKEN="<yourUsername>:<yourPassword>"

post_json() {
  local path="$1"
  local body="$2"
  curl -X POST \
    "$MILVUS_REST_BASE_URL$path" \
    -H "Authorization: Bearer $MILVUS_AUTH_TOKEN" \
    -H "Content-Type: application/json" \
    -d "$body"
}

BODY=$(cat <<JSON
{
  "model_name": "qwen3.7-max",
  "texts": ["Milvus is an open-source vector database that provides high-performance approximate nearest neighbor search. It is commonly used in RAG, recommendation systems, and image search."],
  "params": {"max_words": 30, "temperature": 0, "enable_thinking": false}
}
JSON
)
RESPONSE_BODY="$(post_json "/v2/vectordb/ai/summarize" "$BODY")"
if command -v jq >/dev/null 2>&1; then
  echo "$RESPONSE_BODY" | jq .
  [ "$(echo "$RESPONSE_BODY" | jq -r '.code // -1')" = "0" ] || exit 1
else
  echo "$RESPONSE_BODY"
fi

    Python

    from __future__ import annotations

from typing import Any

from pymilvus import DataType, Function, FunctionType, MilvusClient

MILVUS_URI = "http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_TOKEN = "<yourUsername>:<yourPassword>"

DUMMY_VECTOR_DIM = 2
TEXTTRANSFORM_FUNCTION_TYPE = 9

def texttransform_function_type() -> Any:
    for type_name in ("TEXTTRANSFORM", "TEXT_TRANSFORM", "TextTransform"):
        function_type = getattr(FunctionType, type_name, None)
        if function_type is not None:
            return function_type
    # 阿里雲 Milvus 將 TEXTTRANSFORM 作為託管擴充(函數類型值 9)提供;
    # 部分 pymilvus 版本尚未內建該枚舉成員,而 Function(...) 通過 FunctionType(...) 校正。
    existing = getattr(FunctionType, "_value2member_map_", {}).get(TEXTTRANSFORM_FUNCTION_TYPE)
    if existing is not None:
        return existing
    extension = int.__new__(FunctionType, TEXTTRANSFORM_FUNCTION_TYPE)
    extension._name_ = "TEXTTRANSFORM"
    extension._value_ = TEXTTRANSFORM_FUNCTION_TYPE
    FunctionType._value2member_map_[TEXTTRANSFORM_FUNCTION_TYPE] = extension
    FunctionType._member_map_["TEXTTRANSFORM"] = extension
    return extension

def add_id(schema: Any) -> None:
    schema.add_field("id", DataType.INT64, is_primary=True)

def add_dummy_vector(schema: Any) -> None:
    schema.add_field("dummy_vector", DataType.FLOAT_VECTOR, dim=DUMMY_VECTOR_DIM)

def run_texttransform_example(*, client, collection_name, input_fields, output_field, function_name, function_params, rows) -> None:
    if client.has_collection(collection_name):
        client.drop_collection(collection_name)
    schema = MilvusClient.create_schema(auto_id=True, enable_dynamic_field=False)
    add_id(schema)
    for name, data_type, max_length in input_fields:
        field_params = {"max_length": max_length} if max_length is not None else {}
        schema.add_field(name, data_type, **field_params)
    output_name, output_data_type, output_max_length = output_field
    output_params = {"max_length": output_max_length} if output_max_length is not None else {}
    schema.add_field(output_name, output_data_type, **output_params)
    add_dummy_vector(schema)
    schema.add_function(
        Function(
            name=function_name,
            function_type=texttransform_function_type(),
            input_field_names=[name for name, _, _ in input_fields],
            output_field_names=[output_name],
            params=function_params,
        )
    )
    index_params = client.prepare_index_params()
    index_params.add_index(field_name="dummy_vector", index_type="AUTOINDEX", metric_type="COSINE")
    client.create_collection(collection_name=collection_name, schema=schema, index_params=index_params)
    client.insert(collection_name, rows)
    client.flush(collection_name)
    fields = [name for name, _, _ in input_fields] + [output_name]
    for row in client.query(collection_name, filter="", output_fields=fields, limit=len(rows)):
        print(row)

client = MilvusClient(uri=MILVUS_URI, token=MILVUS_TOKEN)

run_texttransform_example(
    client=client,
    collection_name="simple_ai_summarize_text",
    input_fields=[("content", DataType.VARCHAR, 8192)],
    output_field=("summary", DataType.VARCHAR, 1024),
    function_name="summarize_content",
    function_params={"provider": "aliyun_milvus", "model_name": "qwen3.7-max", "task": "ai_summarize", "max_words": "30", "temperature": "0", "enable_thinking": "false"},
    rows=[{"content": "Milvus is an open-source vector database for high-performance approximate nearest neighbor search in RAG, recommendation, and image search.", "dummy_vector": [0.1, 0.2]}],
)

    預期結果：summary 得到一句約 30 詞的英文導讀，概括 Milvus 的定位與用途。

    樣本二：圖片內容產生摘要

    內容團隊需要為商品圖片產生無障礙描述和檢索摘要，使用多模態模型 qwen3.7-plus。

    REST 介面

    #!/usr/bin/env bash
set -euo pipefail

MILVUS_REST_BASE_URL="http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_AUTH_TOKEN="<yourUsername>:<yourPassword>"

post_json() {
  local path="$1"
  local body="$2"
  curl -X POST \
    "$MILVUS_REST_BASE_URL$path" \
    -H "Authorization: Bearer $MILVUS_AUTH_TOKEN" \
    -H "Content-Type: application/json" \
    -d "$body"
}

BODY=$(cat <<JSON
{"model_name":"qwen3.7-plus","texts":["https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260415/hynnff/wan-video-edit-clothes.webp"],"params":{"media_type":"image","max_words":30,"prompt":"Summarize the main visual content.","temperature":0}}
JSON
)
RESPONSE_BODY="$(post_json "/v2/vectordb/ai/summarize" "$BODY")"
if command -v jq >/dev/null 2>&1; then
  echo "$RESPONSE_BODY" | jq .
  [ "$(echo "$RESPONSE_BODY" | jq -r '.code // -1')" = "0" ] || exit 1
else
  echo "$RESPONSE_BODY"
fi

    Python

    from __future__ import annotations

from typing import Any

from pymilvus import DataType, Function, FunctionType, MilvusClient

MILVUS_URI = "http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_TOKEN = "<yourUsername>:<yourPassword>"

DUMMY_VECTOR_DIM = 2
TEXTTRANSFORM_FUNCTION_TYPE = 9

def texttransform_function_type() -> Any:
    for type_name in ("TEXTTRANSFORM", "TEXT_TRANSFORM", "TextTransform"):
        function_type = getattr(FunctionType, type_name, None)
        if function_type is not None:
            return function_type
    # 阿里雲 Milvus 將 TEXTTRANSFORM 作為託管擴充(函數類型值 9)提供;
    # 部分 pymilvus 版本尚未內建該枚舉成員,而 Function(...) 通過 FunctionType(...) 校正。
    existing = getattr(FunctionType, "_value2member_map_", {}).get(TEXTTRANSFORM_FUNCTION_TYPE)
    if existing is not None:
        return existing
    extension = int.__new__(FunctionType, TEXTTRANSFORM_FUNCTION_TYPE)
    extension._name_ = "TEXTTRANSFORM"
    extension._value_ = TEXTTRANSFORM_FUNCTION_TYPE
    FunctionType._value2member_map_[TEXTTRANSFORM_FUNCTION_TYPE] = extension
    FunctionType._member_map_["TEXTTRANSFORM"] = extension
    return extension

def add_id(schema: Any) -> None:
    schema.add_field("id", DataType.INT64, is_primary=True)

def add_dummy_vector(schema: Any) -> None:
    schema.add_field("dummy_vector", DataType.FLOAT_VECTOR, dim=DUMMY_VECTOR_DIM)

def run_texttransform_example(*, client, collection_name, input_fields, output_field, function_name, function_params, rows) -> None:
    if client.has_collection(collection_name):
        client.drop_collection(collection_name)
    schema = MilvusClient.create_schema(auto_id=True, enable_dynamic_field=False)
    add_id(schema)
    for name, data_type, max_length in input_fields:
        field_params = {"max_length": max_length} if max_length is not None else {}
        schema.add_field(name, data_type, **field_params)
    output_name, output_data_type, output_max_length = output_field
    output_params = {"max_length": output_max_length} if output_max_length is not None else {}
    schema.add_field(output_name, output_data_type, **output_params)
    add_dummy_vector(schema)
    schema.add_function(
        Function(
            name=function_name,
            function_type=texttransform_function_type(),
            input_field_names=[name for name, _, _ in input_fields],
            output_field_names=[output_name],
            params=function_params,
        )
    )
    index_params = client.prepare_index_params()
    index_params.add_index(field_name="dummy_vector", index_type="AUTOINDEX", metric_type="COSINE")
    client.create_collection(collection_name=collection_name, schema=schema, index_params=index_params)
    client.insert(collection_name, rows)
    client.flush(collection_name)
    fields = [name for name, _, _ in input_fields] + [output_name]
    for row in client.query(collection_name, filter="", output_fields=fields, limit=len(rows)):
        print(row)

client = MilvusClient(uri=MILVUS_URI, token=MILVUS_TOKEN)

run_texttransform_example(
    client=client,
    collection_name="simple_ai_summarize_image",
    input_fields=[("image_url", DataType.VARCHAR, 4096)],
    output_field=("summary", DataType.VARCHAR, 1024),
    function_name="summarize_image",
    function_params={"provider": "aliyun_milvus", "model_name": "qwen3.7-plus", "task": "ai_summarize", "media_type": "image", "max_words": "30", "prompt": "Summarize the main visual content.", "temperature": "0"},
    rows=[{"image_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260415/hynnff/wan-video-edit-clothes.webp", "dummy_vector": [0.1, 0.2]}],
)

    預期結果：返回一條不超過 30 詞的圖片內容摘要，並寫入 summary。

    樣本三：視頻內容產生摘要

    視頻庫需要在上傳時自動建置事件概要，供營運快速探索，使用多模態模型 qwen3.7-plus。

    REST 介面

    #!/usr/bin/env bash
set -euo pipefail

MILVUS_REST_BASE_URL="http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_AUTH_TOKEN="<yourUsername>:<yourPassword>"

post_json() {
  local path="$1"
  local body="$2"
  curl -X POST \
    "$MILVUS_REST_BASE_URL$path" \
    -H "Authorization: Bearer $MILVUS_AUTH_TOKEN" \
    -H "Content-Type: application/json" \
    -d "$body"
}

BODY=$(cat <<JSON
{"model_name":"qwen3.7-plus","texts":["https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260409/dozxak/Wan_Video_Edit_33_1.mp4"],"params":{"media_type":"video","max_words":50,"prompt":"Summarize the main events in the video.","temperature":0}}
JSON
)
RESPONSE_BODY="$(post_json "/v2/vectordb/ai/summarize" "$BODY")"
if command -v jq >/dev/null 2>&1; then
  echo "$RESPONSE_BODY" | jq .
  [ "$(echo "$RESPONSE_BODY" | jq -r '.code // -1')" = "0" ] || exit 1
else
  echo "$RESPONSE_BODY"
fi

    Python

    from __future__ import annotations

from typing import Any

from pymilvus import DataType, Function, FunctionType, MilvusClient

MILVUS_URI = "http://c-xxxx.milvus.aliyuncs.com:19530"
MILVUS_TOKEN = "<yourUsername>:<yourPassword>"

DUMMY_VECTOR_DIM = 2
TEXTTRANSFORM_FUNCTION_TYPE = 9

def texttransform_function_type() -> Any:
    for type_name in ("TEXTTRANSFORM", "TEXT_TRANSFORM", "TextTransform"):
        function_type = getattr(FunctionType, type_name, None)
        if function_type is not None:
            return function_type
    # 阿里雲 Milvus 將 TEXTTRANSFORM 作為託管擴充(函數類型值 9)提供;
    # 部分 pymilvus 版本尚未內建該枚舉成員,而 Function(...) 通過 FunctionType(...) 校正。
    existing = getattr(FunctionType, "_value2member_map_", {}).get(TEXTTRANSFORM_FUNCTION_TYPE)
    if existing is not None:
        return existing
    extension = int.__new__(FunctionType, TEXTTRANSFORM_FUNCTION_TYPE)
    extension._name_ = "TEXTTRANSFORM"
    extension._value_ = TEXTTRANSFORM_FUNCTION_TYPE
    FunctionType._value2member_map_[TEXTTRANSFORM_FUNCTION_TYPE] = extension
    FunctionType._member_map_["TEXTTRANSFORM"] = extension
    return extension

def add_id(schema: Any) -> None:
    schema.add_field("id", DataType.INT64, is_primary=True)

def add_dummy_vector(schema: Any) -> None:
    schema.add_field("dummy_vector", DataType.FLOAT_VECTOR, dim=DUMMY_VECTOR_DIM)

def run_texttransform_example(*, client, collection_name, input_fields, output_field, function_name, function_params, rows) -> None:
    if client.has_collection(collection_name):
        client.drop_collection(collection_name)
    schema = MilvusClient.create_schema(auto_id=True, enable_dynamic_field=False)
    add_id(schema)
    for name, data_type, max_length in input_fields:
        field_params = {"max_length": max_length} if max_length is not None else {}
        schema.add_field(name, data_type, **field_params)
    output_name, output_data_type, output_max_length = output_field
    output_params = {"max_length": output_max_length} if output_max_length is not None else {}
    schema.add_field(output_name, output_data_type, **output_params)
    add_dummy_vector(schema)
    schema.add_function(
        Function(
            name=function_name,
            function_type=texttransform_function_type(),
            input_field_names=[name for name, _, _ in input_fields],
            output_field_names=[output_name],
            params=function_params,
        )
    )
    index_params = client.prepare_index_params()
    index_params.add_index(field_name="dummy_vector", index_type="AUTOINDEX", metric_type="COSINE")
    client.create_collection(collection_name=collection_name, schema=schema, index_params=index_params)
    client.insert(collection_name, rows)
    client.flush(collection_name)
    fields = [name for name, _, _ in input_fields] + [output_name]
    for row in client.query(collection_name, filter="", output_fields=fields, limit=len(rows)):
        print(row)

client = MilvusClient(uri=MILVUS_URI, token=MILVUS_TOKEN)

run_texttransform_example(
    client=client,
    collection_name="simple_ai_summarize_video",
    input_fields=[("video_url", DataType.VARCHAR, 4096)],
    output_field=("summary", DataType.VARCHAR, 1024),
    function_name="summarize_video",
    function_params={"provider": "aliyun_milvus", "model_name": "qwen3.7-plus", "task": "ai_summarize", "media_type": "video", "max_words": "50", "prompt": "Summarize the main events in the video.", "temperature": "0"},
    rows=[{"video_url": "https://help-static-aliyun-doc.aliyuncs.com/file-manage-files/zh-CN/20260409/dozxak/Wan_Video_Edit_33_1.mp4", "dummy_vector": [0.1, 0.2]}],
)

    預期結果：返回視頻主要事件的簡潔摘要，並寫入 summary。