教程:Gemini API 使用入门


在 ai.google.dev 上查看 在 Google Colab 中运行 在 GitHub 上查看源代码

如果您想快速试用 Gemini API,可以使用 curl 命令调用 REST API 中的方法。本教程中的示例展示了对每种 API 方法的调用。

Colab 使用 Python 代码来设置环境变量并显示图片,但您不需要 Colab 也能使用 REST API。只要您按照下一部分中的说明设置了 API_KEY,就无需修改即可在 Colab 外部运行所有 curl 示例。

对于每个 curl 命令,您都必须指定适用的模型名称和 API 密钥。

设置您的 API 密钥

如需使用 Gemini API,您需要 API 密钥。如果您还没有密钥,请在 Google AI Studio 中创建一个。

获取 API 密钥

在 Colab 中,将密钥添加到左侧面板“🔑?”下的 Secret 管理器中。将其命名为 GOOGLE_API_KEY。然后,您可以将其添加为环境变量,以便在 curl 调用中传递密钥。

在终端中,您只需运行 GOOGLE_API_KEY="Your API Key"

import os
from google.colab import userdata

os.environ['GOOGLE_API_KEY'] = userdata.get('GOOGLE_API_KEY')

基于 Gemini 和 Content 的 API

纯文字输入

使用 generateContent 方法在给定输入消息的情况下根据模型生成回复。如果输入仅包含文本,请使用 Gemini 1.5 模型或 Gemini 1.0 Pro 模型。

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
      "contents": [{
        "parts":[{
          "text": "Write a story about a magic backpack."}]}]}' 2> /dev/null
{
  "candidates": [
    {
      "content": {
        "parts": [
          {
            "text": "In the quaint town of Willow Creek, nestled amidst rolling hills and whispering willows, there lived an ordinary boy named Ethan. Ethan's life took an extraordinary turn the day he stumbled upon an enigmatic backpack hidden in the depths of his attic.\n\nCuriosity ignited within Ethan as he lifted the worn leather straps and unzipped its mysterious contents. Inside lay a shimmering array of vibrant objects and peculiar trinkets. There was a glowing orb that pulsated with an ethereal glow, a feather that seemed to have a life of its own, and a small, enigmatic key.\n\nAs Ethan explored each item, he realized they possessed astonishing abilities. The orb illuminated his path, casting a warm glow in the darkest of nights. The feather granted him the power of flight, allowing him to soar through the skies with newfound freedom. And the key opened a portal to a hidden world, a realm of endless wonder.\n\nArmed with his magical backpack, Ethan embarked on countless adventures. He flew over the towering mountains of Willow Creek, exploring their hidden secrets. He navigated the treacherous depths of the Enchanted Forest, where he encountered mythical creatures and ancient spirits. And he ventured into distant, unknown lands, uncovering lost civilizations and forgotten treasures.\n\nWith each adventure, Ethan's knowledge and abilities grew. He learned to harness the power of his backpack wisely, using its magic to help others and protect the world from evil forces. The backpack became an extension of himself, a symbol of hope and wonder in the face of adversity.\n\nAs the years went by, Ethan's reputation as the boy with the magic backpack spread far and wide. People from all walks of life came to him, seeking his guidance and protection. And Ethan never hesitated to lend a helping hand, using his extraordinary abilities to make the world a better place.\n\nIn the end, the magic backpack became more than just a collection of objects. It was a representation of Ethan's unwavering spirit, his boundless imagination, and his unwavering belief in the power of dreams. And as long as Ethan carried it with him, the magic of Willow Creek would live on, illuminating the darkest corners of the world with hope, wonder, and the limitless possibilities that resided within the heart of a child."
          }
        ],
        "role": "model"
      },
      "finishReason": "STOP",
      "index": 0,
      "safetyRatings": [
        {
          "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_HATE_SPEECH",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_HARASSMENT",
          "probability": "NEGLIGIBLE"
        },
        {
          "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
          "probability": "NEGLIGIBLE"
        }
      ]
    }
  ],
  "promptFeedback": {
    "safetyRatings": [
      {
        "category": "HARM_CATEGORY_SEXUALLY_EXPLICIT",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_HATE_SPEECH",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_HARASSMENT",
        "probability": "NEGLIGIBLE"
      },
      {
        "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
        "probability": "NEGLIGIBLE"
      }
    ]
  }
}

文字和图片输入

Gemini 提供了多种可处理多模态输入的模型(Gemini 1.5 模型),让您可以输入文字和图片。请务必查看提示的图片要求

如果输入的提示同时包含文本和图片,请使用 Gemini 1.5 模型。

以下代码段可帮助您构建请求并将其发送到 REST API。

curl -o image.jpg https://storage.googleapis.com/generativeai-downloads/images/scones.jpg
% Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed
100  385k  100  385k    0     0  2053k      0 --:--:-- --:--:-- --:--:-- 2050k
import PIL.Image

img = PIL.Image.open("image.jpg")
img.resize((512, int(img.height*512/img.width)))

png

echo '{
  "contents":[
    {
      "parts":[
        {"text": "What is this picture?"},
        {
          "inline_data": {
            "mime_type":"image/jpeg",
            "data": "'$(base64 -w0 image.jpg)'"
          }
        }
      ]
    }
  ]
}' > request.json
curl https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent?key=${GOOGLE_API_KEY} \
        -H 'Content-Type: application/json' \
        -d @request.json 2> /dev/null | grep "text"
"text": " The picture shows a table with a white tablecloth. On the table are two cups of coffee, a bowl of blueberries, and a plate of scones. There are also some flowers on the table."

多轮对话(聊天)

借助 Gemini,您可以跨多个回合构建自由形式的对话。

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
      "contents": [
        {"role":"user",
         "parts":[{
           "text": "Write the first line of a story about a magic backpack."}]},
        {"role": "model",
         "parts":[{
           "text": "In the bustling city of Meadow brook, lived a young girl named Sophie. She was a bright and curious soul with an imaginative mind."}]},
        {"role": "user",
         "parts":[{
           "text": "Can you set it in a quiet village in 1600s France?"}]},
      ]
    }' 2> /dev/null | grep "text"
"text": "In the quaint village of Fleur-de-Lys, nestled amidst the rolling hills of 17th century France, lived a young maiden named Antoinette. She possessed a heart brimming with curiosity and a spirit as vibrant as the wildflowers that bloomed in the meadows.\n\nOne sunny morn, as Antoinette strolled through the cobblestone streets, her gaze fell upon a peculiar sight—a weathered leather backpack resting atop a mossy stone bench. Intrigued, she cautiously approached the bag, her fingers tracing the intricate carvings etched into its surface. As her fingertips grazed the worn leather, a surge of warmth coursed through her body, and the backpack began to emit a soft, ethereal glow."

配置

您发送到模型的每个提示都包含参数值,用于控制模型如何生成回答。对于不同的参数值,模型会生成不同的结果。详细了解模型参数

此外,您还可以使用安全设置来调整收到可能有害回答的可能性。默认情况下,安全设置会在所有维度上屏蔽不安全内容的中等和/或高概率。不妨详细了解安全设置

以下示例指定了 generateContent 方法的所有参数的值。

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
        "contents": [{
            "parts":[
                {"text": "Write a story about a magic backpack."}
            ]
        }],
        "safetySettings": [
            {
                "category": "HARM_CATEGORY_DANGEROUS_CONTENT",
                "threshold": "BLOCK_ONLY_HIGH"
            }
        ],
        "generationConfig": {
            "stopSequences": [
                "Title"
            ],
            "temperature": 1.0,
            "maxOutputTokens": 800,
            "topP": 0.8,
            "topK": 10
        }
    }'  2> /dev/null | grep "text"
"text": "Once upon a time, in a small town nestled at the foot of a majestic mountain range, lived a young girl named Lily. Lily was a bright and curious child who loved to explore the world around her. One day, while playing in the forest near her home, she stumbled upon a hidden cave. Intrigued, she stepped inside, and to her amazement, she discovered a dusty old backpack lying in a corner.\n\nCuriosity piqued, Lily reached out and picked up the backpack. As soon as her fingers brushed against the worn leather, she felt a strange tingling sensation coursing through her body. Suddenly, the backpack began to glow, emitting a soft, ethereal light that filled the cave.\n\nWith wide-eyed wonder, Lily opened the backpack to find it filled with an assortment of magical objects. There was a compass that always pointed to the nearest adventure, a magnifying glass that could reveal hidden secrets, a telescope that allowed her to see distant lands, and a book that contained the knowledge of the universe.\n\nOverjoyed with her discovery, Lily took the magic backpack home and began using its contents to explore the world in ways she had never imagined. She followed the compass to discover hidden treasures, used the magnifying glass to uncover the secrets of nature, gazed through the telescope to witness the wonders of the cosmos, and delved into the book to learn about the mysteries of the universe.\n\nAs Lily's adventures continued, she realized that the magic backpack was more than just a collection of enchanted items. It was a symbol of her own limitless potential and the power of her imagination. It taught her that with curiosity, courage, and a touch of magic, anything was possible.\n\nNews of Lily's magical backpack spread throughout the town, and soon, children from all around came to her, eager to learn about its wonders. Lily welcomed them with open arms, sharing her stories and inspiring them to embark on their own adventures.\n\nAnd so, the magic backpack became a beacon of hope and wonder, reminding everyone that the world is full of hidden treasures waiting to be discovered, if only one has the courage to step into the unknown."

直播生成内容

generateContent 方法会在完成整个生成过程后返回响应。通过不等待整个结果,您可以实现更快的互动,并改用 streamGenerateContent 返回部分结果。

!curl "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:streamGenerateContent?alt=sse&key=${GOOGLE_API_KEY}" \
        -H 'Content-Type: application/json' \
        --no-buffer \
        -d '{ "contents":[{"parts":[{"text": "Write long a story about a magic backpack."}]}]}' \
        2> /dev/null
data: {"candidates": [{"content": {"parts": [{"text": "In the quaint little town of Willow Creek, nestled among rolling hills and whispering willows"}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}],"promptFeedback": {"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]} }

data: {"candidates": [{"content": {"parts": [{"text": ", there existed an extraordinary backpack that possessed an astonishing secret. Its unassuming canvas exterior and worn leather straps held a hidden realm brimming with wonder and endless possibilities."}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}]}

data: {"candidates": [{"content": {"parts": [{"text": "\n\nYoung Oliver, a curious and imaginative boy, stumbled upon this magical backpack in the dusty attic of his grandmother's house. Intrigued by its enigmatic aura, he unzipped it cautiously, revealing a seemingly ordinary interior. But as his fingers brushed against the lining, an ethereal glow emanated from within.\n\n"}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}]}

data: {"candidates": [{"content": {"parts": [{"text": "With a gasp of surprise, Oliver watched as the backpack transformed before his very eyes. Its fabric shimmered and flowed like liquid silver, morphing into a portal that connected him to a hidden dimension. Step by step, he ventured into this enchanted realm, his heart pounding with a mixture of trepidation and exhilaration.\n\nThe backpack's interior was a vast and wondrous labyrinth filled with towering bookshelves, bubbling potions, and ethereal artifacts. Each turn offered a new discovery: a self-playing piano, a talking mirror that whispered ancient wisdom, and a compass that pointed to the furthest reaches of the imagination.\n\nOliver soon realized"}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}]}

data: {"candidates": [{"content": {"parts": [{"text": " that this backpack was no mere container but a sentient being, capable of aiding him in his quests and expanding his horizons. It granted him the gift of tongues, allowing him to speak with animals and creatures from distant lands. It gifted him a quill that wrote stories that danced off the page, bringing his wildest dreams to life.\n\nTogether, Oliver and the backpack embarked on extraordinary adventures. They soared through the skies on the back of a majestic griffon, traversed treacherous terrains with the aid of a shape-shifting fox, and solved mysteries that had long baffled the wisest minds in Willow Creek.\n\nAs the days turned into weeks, Oliver's imagination flourished beyond measure. He painted vibrant landscapes with words, composed symphonies that echoed through the hidden realm, and invented gadgets that defied the laws of physics. The backpack became an extension of his boundless creativity, nurturing his wonder and fueling his aspirations.\n\nNews of Oliver's extraordinary backpack spread throughout the town and beyond. People flocked from far and wide to witness its marvels. Scholars sought its wisdom, artists sought its inspiration, and children dreamed of experiencing its boundless adventures.\n\nHowever, not all who approached the backpack with pure intentions. One fateful day, a greedy sorcerer named Maldred attempted to seize its power for himself. But"}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}]}

data: {"candidates": [{"content": {"parts": [{"text": " the backpack, sensing his malevolent nature, resisted his grasp and summoned a legion of fantastical creatures to its defense.\n\nIn a fierce battle that shook the very fabric of the hidden realm, Oliver and the backpack allied with brave heroes and wise wizards to defeat Maldred and his wicked forces. The town of Willow Creek was forever grateful, and the backpack became a symbol of hope and imagination for all who knew of its existence.\n\nAs the years passed, Oliver grew into a wise and compassionate leader, using the magic backpack to spread joy, inspire creativity, and unlock the potential of those around him. The hidden realm within its depths became a sanctuary for dreamers, inventors, and anyone who dared to embrace the wonders of the unknown.\n\nAnd so, the tale of the magic backpack was passed down through generations, a timeless testament to the power of imagination and the boundless possibilities that lie when wonder and curiosity ignite the human spirit."}],"role": "model"},"finishReason": "STOP","index": 0,"safetyRatings": [{"category": "HARM_CATEGORY_SEXUALLY_EXPLICIT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HATE_SPEECH","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_HARASSMENT","probability": "NEGLIGIBLE"},{"category": "HARM_CATEGORY_DANGEROUS_CONTENT","probability": "NEGLIGIBLE"}]}]}

计算词元数量

使用长提示时,在向模型发送任何内容之前统计词元数量可能会很有用。

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:countTokens?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
      "contents": [{
        "parts":[{
          "text": "Write a story about a magic backpack."}]}]}' > response.json
% Total    % Received % Xferd  Average Speed   Time    Time     Time  Current
                                 Dload  Upload   Total   Spent    Left  Speed

  0     0    0     0    0     0      0      0 --:--:-- --:--:-- --:--:--     0
100   127    0    23  100   104    105    477 --:--:-- --:--:-- --:--:--   585
cat response.json
{
  "totalTokens": 8
}

嵌入

嵌入是一种技术,用于将信息表示为数组中的浮点数列表。借助 Gemini,您能够以矢量化形式表示文本(字词、句子和文本块),从而更轻松地比较和对比嵌入。例如,主题或情感相似的两个文本应该具有相似的嵌入,这些嵌入可以通过余弦相似度等数学比较技术来识别。

embedding-001 模型与 embedContentsbatchEmbedContents 搭配使用:

curl https://generativelanguage.googleapis.com/v1beta/models/embedding-001:embedContent?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
        "model": "models/embedding-001",
        "content": {
        "parts":[{
          "text": "Write a story about a magic backpack."}]} }' 2> /dev/null | head
{
  "embedding": {
    "values": [
      0.008624583,
      -0.030451821,
      -0.042496547,
      -0.029230341,
      0.05486475,
      0.006694871,
      0.004025645,
curl https://generativelanguage.googleapis.com/v1beta/models/embedding-001:batchEmbedContents?key=$GOOGLE_API_KEY \
    -H 'Content-Type: application/json' \
    -X POST \
    -d '{
      "requests": [{
        "model": "models/embedding-001",
        "content": {
        "parts":[{
          "text": "Write a story about a magic backpack."}]} }]}' 2> /dev/null | head
{
  "embeddings": [
    {
      "values": [
        0.008624583,
        -0.030451821,
        -0.042496547,
        -0.029230341,
        0.05486475,
        0.006694871,

型号信息

获取模型

如果使用 GET 模型的网址,API 会使用 get 方法返回有关该模型的信息,例如版本、显示名称、输入词元限制等。

curl https://generativelanguage.googleapis.com/v1beta/models/gemini-pro?key=$GOOGLE_API_KEY
{
  "name": "models/gemini-pro",
  "version": "001",
  "displayName": "Gemini Pro",
  "description": "The best model for scaling across a wide range of tasks",
  "inputTokenLimit": 30720,
  "outputTokenLimit": 2048,
  "supportedGenerationMethods": [
    "generateContent",
    "countTokens"
  ],
  "temperature": 0.9,
  "topP": 1,
  "topK": 1
}

列出模型

如果您 GETmodels 目录,它会使用 list 方法列出该 API 提供的所有模型,包括 Gemini 和 PaLM 系列模型。

curl https://generativelanguage.googleapis.com/v1beta/models?key=$GOOGLE_API_KEY
{
  "models": [
    {
      "name": "models/chat-bison-001",
      "version": "001",
      "displayName": "Chat Bison",
      "description": "Chat-optimized generative language model.",
      "inputTokenLimit": 4096,
      "outputTokenLimit": 1024,
      "supportedGenerationMethods": [
        "generateMessage",
        "countMessageTokens"
      ],
      "temperature": 0.25,
      "topP": 0.95,
      "topK": 40
    },
    {
      "name": "models/text-bison-001",
      "version": "001",
      "displayName": "Text Bison",
      "description": "Model targeted for text generation.",
      "inputTokenLimit": 8196,
      "outputTokenLimit": 1024,
      "supportedGenerationMethods": [
        "generateText",
        "countTextTokens",
        "createTunedTextModel"
      ],
      "temperature": 0.7,
      "topP": 0.95,
      "topK": 40
    },
    {
      "name": "models/embedding-gecko-001",
      "version": "001",
      "displayName": "Embedding Gecko",
      "description": "Obtain a distributed representation of a text.",
      "inputTokenLimit": 1024,
      "outputTokenLimit": 1,
      "supportedGenerationMethods": [
        "embedText",
        "countTextTokens"
      ]
    },
    {
      "name": "models/embedding-gecko-002",
      "version": "002",
      "displayName": "Embedding Gecko 002",
      "description": "Obtain a distributed representation of a text.",
      "inputTokenLimit": 2048,
      "outputTokenLimit": 1,
      "supportedGenerationMethods": [
        "embedText",
        "countTextTokens"
      ]
    },
    {
      "name": "models/gemini-pro",
      "version": "001",
      "displayName": "Gemini Pro",
      "description": "The best model for scaling across a wide range of tasks",
      "inputTokenLimit": 30720,
      "outputTokenLimit": 2048,
      "supportedGenerationMethods": [
        "generateContent",
        "countTokens"
      ],
      "temperature": 0.9,
      "topP": 1,
      "topK": 1
    },
    {
      "name": "models/gemini-pro-vision",
      "version": "001",
      "displayName": "Gemini Pro Vision",
      "description": "The best image understanding model to handle a broad range of applications",
      "inputTokenLimit": 12288,
      "outputTokenLimit": 4096,
      "supportedGenerationMethods": [
        "generateContent",
        "countTokens"
      ],
      "temperature": 0.4,
      "topP": 1,
      "topK": 32
    },
    {
      "name": "models/gemini-ultra",
      "version": "001",
      "displayName": "Gemini Ultra",
      "description": "The most capable model for highly complex tasks",
      "inputTokenLimit": 30720,
      "outputTokenLimit": 2048,
      "supportedGenerationMethods": [
        "generateContent",
        "countTokens"
      ],
      "temperature": 0.9,
      "topP": 1,
      "topK": 32
    },
    {
      "name": "models/embedding-001",
      "version": "001",
      "displayName": "Embedding 001",
      "description": "Obtain a distributed representation of a text.",
      "inputTokenLimit": 2048,
      "outputTokenLimit": 1,
      "supportedGenerationMethods": [
        "embedContent",
        "countTextTokens"
      ]
    },
    {
      "name": "models/aqa",
      "version": "001",
      "displayName": "Model that performs Attributed Question Answering.",
      "description": "Model trained to return answers to questions that are grounded in provided sources, along with estimating answerable probability.",
      "inputTokenLimit": 7168,
      "outputTokenLimit": 1024,
      "supportedGenerationMethods": [
        "generateAnswer"
      ],
      "temperature": 0.2,
      "topP": 1,
      "topK": 40
    }
  ]
}