Gemini 图像生成 API

curl --request POST \
  --url https://www.geeknow.top/v1beta/models/{model}:generateContent \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "contents": [
    {}
  ],
  "contents[].role": "<string>",
  "contents[].parts": [
    {}
  ],
  "contents[].parts[].text": "<string>",
  "contents[].parts[].inlineData": {},
  "contents[].parts[].inlineData.mimeType": "<string>",
  "contents[].parts[].inlineData.data": "<string>",
  "generationConfig": {},
  "generationConfig.responseModalities": [
    "<string>"
  ],
  "generationConfig.imageConfig": {},
  "generationConfig.imageConfig.aspectRatio": "<string>",
  "generationConfig.imageConfig.imageSize": "<string>",
  "safetySettings": [
    {}
  ]
}
'

import requests

url = "https://www.geeknow.top/v1beta/models/{model}:generateContent"

payload = {
    "contents": [{}],
    "contents[].role": "<string>",
    "contents[].parts": [{}],
    "contents[].parts[].text": "<string>",
    "contents[].parts[].inlineData": {},
    "contents[].parts[].inlineData.mimeType": "<string>",
    "contents[].parts[].inlineData.data": "<string>",
    "generationConfig": {},
    "generationConfig.responseModalities": ["<string>"],
    "generationConfig.imageConfig": {},
    "generationConfig.imageConfig.aspectRatio": "<string>",
    "generationConfig.imageConfig.imageSize": "<string>",
    "safetySettings": [{}]
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    contents: [{}],
    'contents[].role': '<string>',
    'contents[].parts': [{}],
    'contents[].parts[].text': '<string>',
    'contents[].parts[].inlineData': {},
    'contents[].parts[].inlineData.mimeType': '<string>',
    'contents[].parts[].inlineData.data': '<string>',
    generationConfig: {},
    'generationConfig.responseModalities': ['<string>'],
    'generationConfig.imageConfig': {},
    'generationConfig.imageConfig.aspectRatio': '<string>',
    'generationConfig.imageConfig.imageSize': '<string>',
    safetySettings: [{}]
  })
};

fetch('https://www.geeknow.top/v1beta/models/{model}:generateContent', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://www.geeknow.top/v1beta/models/{model}:generateContent",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'contents' => [
        [
                
        ]
    ],
    'contents[].role' => '<string>',
    'contents[].parts' => [
        [
                
        ]
    ],
    'contents[].parts[].text' => '<string>',
    'contents[].parts[].inlineData' => [
        
    ],
    'contents[].parts[].inlineData.mimeType' => '<string>',
    'contents[].parts[].inlineData.data' => '<string>',
    'generationConfig' => [
        
    ],
    'generationConfig.responseModalities' => [
        '<string>'
    ],
    'generationConfig.imageConfig' => [
        
    ],
    'generationConfig.imageConfig.aspectRatio' => '<string>',
    'generationConfig.imageConfig.imageSize' => '<string>',
    'safetySettings' => [
        [
                
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://www.geeknow.top/v1beta/models/{model}:generateContent"

	payload := strings.NewReader("{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://www.geeknow.top/v1beta/models/{model}:generateContent")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://www.geeknow.top/v1beta/models/{model}:generateContent")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}"

response = http.request(request)
puts response.read_body

{
  "candidates": [
    {
      "content": {
        "role": "model",
        "parts": [
          {
            "inlineData": {
              "mimeType": "image/png",
              "data": "BASE64_OR_URL"
            }
          }
        ]
      },
      "finishReason": "STOP"
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 123,
    "candidatesTokenCount": 456,
    "totalTokenCount": 579,
    "trafficType": "ON_DEMAND"
  },
  "modelVersion": "gemini-3-pro-image-preview",
  "createTime": "2026-05-21T00:00:00Z"
}

POST

https://www.geeknow.top

v1beta

models

{model}

:generateContent

Gemini 图像生成 API

curl --request POST \
  --url https://www.geeknow.top/v1beta/models/{model}:generateContent \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "contents": [
    {}
  ],
  "contents[].role": "<string>",
  "contents[].parts": [
    {}
  ],
  "contents[].parts[].text": "<string>",
  "contents[].parts[].inlineData": {},
  "contents[].parts[].inlineData.mimeType": "<string>",
  "contents[].parts[].inlineData.data": "<string>",
  "generationConfig": {},
  "generationConfig.responseModalities": [
    "<string>"
  ],
  "generationConfig.imageConfig": {},
  "generationConfig.imageConfig.aspectRatio": "<string>",
  "generationConfig.imageConfig.imageSize": "<string>",
  "safetySettings": [
    {}
  ]
}
'

import requests

url = "https://www.geeknow.top/v1beta/models/{model}:generateContent"

payload = {
    "contents": [{}],
    "contents[].role": "<string>",
    "contents[].parts": [{}],
    "contents[].parts[].text": "<string>",
    "contents[].parts[].inlineData": {},
    "contents[].parts[].inlineData.mimeType": "<string>",
    "contents[].parts[].inlineData.data": "<string>",
    "generationConfig": {},
    "generationConfig.responseModalities": ["<string>"],
    "generationConfig.imageConfig": {},
    "generationConfig.imageConfig.aspectRatio": "<string>",
    "generationConfig.imageConfig.imageSize": "<string>",
    "safetySettings": [{}]
}
headers = {
    "Authorization": "Bearer <token>",
    "Content-Type": "application/json"
}

response = requests.post(url, json=payload, headers=headers)

print(response.text)

const options = {
  method: 'POST',
  headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
  body: JSON.stringify({
    contents: [{}],
    'contents[].role': '<string>',
    'contents[].parts': [{}],
    'contents[].parts[].text': '<string>',
    'contents[].parts[].inlineData': {},
    'contents[].parts[].inlineData.mimeType': '<string>',
    'contents[].parts[].inlineData.data': '<string>',
    generationConfig: {},
    'generationConfig.responseModalities': ['<string>'],
    'generationConfig.imageConfig': {},
    'generationConfig.imageConfig.aspectRatio': '<string>',
    'generationConfig.imageConfig.imageSize': '<string>',
    safetySettings: [{}]
  })
};

fetch('https://www.geeknow.top/v1beta/models/{model}:generateContent', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

<?php

$curl = curl_init();

curl_setopt_array($curl, [
  CURLOPT_URL => "https://www.geeknow.top/v1beta/models/{model}:generateContent",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_POSTFIELDS => json_encode([
    'contents' => [
        [
                
        ]
    ],
    'contents[].role' => '<string>',
    'contents[].parts' => [
        [
                
        ]
    ],
    'contents[].parts[].text' => '<string>',
    'contents[].parts[].inlineData' => [
        
    ],
    'contents[].parts[].inlineData.mimeType' => '<string>',
    'contents[].parts[].inlineData.data' => '<string>',
    'generationConfig' => [
        
    ],
    'generationConfig.responseModalities' => [
        '<string>'
    ],
    'generationConfig.imageConfig' => [
        
    ],
    'generationConfig.imageConfig.aspectRatio' => '<string>',
    'generationConfig.imageConfig.imageSize' => '<string>',
    'safetySettings' => [
        [
                
        ]
    ]
  ]),
  CURLOPT_HTTPHEADER => [
    "Authorization: Bearer <token>",
    "Content-Type: application/json"
  ],
]);

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

package main

import (
	"fmt"
	"strings"
	"net/http"
	"io"
)

func main() {

	url := "https://www.geeknow.top/v1beta/models/{model}:generateContent"

	payload := strings.NewReader("{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}")

	req, _ := http.NewRequest("POST", url, payload)

	req.Header.Add("Authorization", "Bearer <token>")
	req.Header.Add("Content-Type", "application/json")

	res, _ := http.DefaultClient.Do(req)

	defer res.Body.Close()
	body, _ := io.ReadAll(res.Body)

	fmt.Println(string(body))

}

HttpResponse<String> response = Unirest.post("https://www.geeknow.top/v1beta/models/{model}:generateContent")
  .header("Authorization", "Bearer <token>")
  .header("Content-Type", "application/json")
  .body("{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}")
  .asString();

require 'uri'
require 'net/http'

url = URI("https://www.geeknow.top/v1beta/models/{model}:generateContent")

http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true

request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n  \"contents\": [\n    {}\n  ],\n  \"contents[].role\": \"<string>\",\n  \"contents[].parts\": [\n    {}\n  ],\n  \"contents[].parts[].text\": \"<string>\",\n  \"contents[].parts[].inlineData\": {},\n  \"contents[].parts[].inlineData.mimeType\": \"<string>\",\n  \"contents[].parts[].inlineData.data\": \"<string>\",\n  \"generationConfig\": {},\n  \"generationConfig.responseModalities\": [\n    \"<string>\"\n  ],\n  \"generationConfig.imageConfig\": {},\n  \"generationConfig.imageConfig.aspectRatio\": \"<string>\",\n  \"generationConfig.imageConfig.imageSize\": \"<string>\",\n  \"safetySettings\": [\n    {}\n  ]\n}"

response = http.request(request)
puts response.read_body

{
  "candidates": [
    {
      "content": {
        "role": "model",
        "parts": [
          {
            "inlineData": {
              "mimeType": "image/png",
              "data": "BASE64_OR_URL"
            }
          }
        ]
      },
      "finishReason": "STOP"
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 123,
    "candidatesTokenCount": 456,
    "totalTokenCount": 579,
    "trafficType": "ON_DEMAND"
  },
  "modelVersion": "gemini-3-pro-image-preview",
  "createTime": "2026-05-21T00:00:00Z"
}

Gemini 图像生成 API

Gemini 图像模型统一使用官方 generateContent 格式生图，适合直接复用 contents / parts 结构。

路径为 POST /v1beta/models/{model}:generateContent。
文生图和图生图都走同一套 contents[].parts[] 结构。
参考图通过 parts[].inlineData 传入。
generationConfig.responseModalities 建议同时请求 IMAGE 和 TEXT。
返回里既可能直接给 inlineData.data Base64，也可能给图片 URL。

支持模型

gemini-3-pro-image-preview
gemini-2.5-flash-image-preview
gemini-3.1-flash-image-preview
gemini-3.1-flash-lite-image

模型差异

模型	`imageSize` 行为	说明
`gemini-3-pro-image-preview`	支持 `1K`、`2K`	会按请求里的 `imageSize` 真实下发
`gemini-2.5-flash-image-preview`	实际回落为 `1K`	即使请求中传 `2K`，最终也会按 `1K` 处理
`gemini-3.1-flash-image-preview`	实际回落为 `1K`	即使请求中传 `2K`，最终也会按 `1K` 处理
`gemini-3.1-flash-lite-image`	实际回落为 `1K`	即使请求中传 `2K`，最终也会按 `1K` 处理

Body

contents

array<object>

required

输入内容列表。每项通常包含 role 和 parts。

contents[].role

string

角色字段，常用值为 user。

contents[].parts

array<object>

required

内容片段数组。提示词和参考图都在这里构造。

contents[].parts[].text

string

文本提示词。

contents[].parts[].inlineData

object

参考图输入。包含 mimeType 和 data 两个字段。

contents[].parts[].inlineData.mimeType

string

图片 MIME 类型，常见值为 image/jpeg、image/png、image/webp。

contents[].parts[].inlineData.data

string

Base64 编码的图片内容。

generationConfig

object

required

生成配置对象。

generationConfig.responseModalities

array<string>

返回模态列表。建议同时请求 IMAGE 和 TEXT。

generationConfig.imageConfig

object

required

图像配置对象。

generationConfig.imageConfig.aspectRatio

string

图片比例。常见值包括 1:1、16:9、9:16、4:3、3:4、3:2、2:3、21:9。

generationConfig.imageConfig.imageSize

string

图片清晰度。gemini-3-pro-image-preview 支持 1K 和 2K；gemini-2.5-flash-image-preview、gemini-3.1-flash-image-preview 与 gemini-3.1-flash-lite-image 会实际回落为 1K。

safetySettings

array<object>

安全设置。可不传。

图生图请求结构

Gemini 不提供单独的编辑接口。图生图请求会将参考图直接作为输入内容写入 contents[].parts[]。可简化理解为：

文生图：parts[] 里只有文本。
图生图：parts[] 里同时包含文本和图片。

也就是说，提示词与参考图会在同一次请求中一并提交给模型。

文生图 vs 图生图

场景	`parts[]` 内容	说明
文生图	只有 `text`	纯提示词生成
图生图	`text` + 一个或多个 `inlineData`	让模型参考已有图片风格、主体或构图

图生图字段构造

请求构造可遵循以下规则：

第一类内容是提示词：{ "text": "..." }
第二类内容是参考图：{ "inlineData": { "mimeType": "...", "data": "BASE64..." } }

构造请求时通常先放文本提示词，再将每一张参考图转成 Base64，按顺序追加到同一个 parts[]。

图生图示例

{
  "contents": [
    {
      "role": "user",
      "parts": [
        { "text": "保留参考图的主体氛围，重做成更高级的海报" },
        {
          "inlineData": {
            "mimeType": "image/jpeg",
            "data": "BASE64_IMAGE"
          }
        }
      ]
    }
  ],
  "generationConfig": {
    "responseModalities": ["TEXT", "IMAGE"],
    "imageConfig": {
      "aspectRatio": "16:9",
      "imageSize": "2K"
    }
  }
}

方法与路径

POST /v1beta/models/{model}:generateContent

请求示例

curl -X POST https://www.geeknow.top/v1beta/models/gemini-3-pro-image-preview:generateContent \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Accept: application/json" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          { "text": "一张未来感 AI 工作台场景，冷色氛围，电影级布光" }
        ]
      }
    ],
    "generationConfig": {
      "responseModalities": ["TEXT", "IMAGE"],
      "temperature": 1.0,
      "topP": 0.95,
      "maxOutputTokens": 8192,
      "imageConfig": {
        "aspectRatio": "16:9",
        "imageSize": "2K"
      }
    }
  }'

import requests

resp = requests.post(
    "https://www.geeknow.top/v1beta/models/gemini-3-pro-image-preview:generateContent",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Accept": "application/json",
        "Content-Type": "application/json",
    },
    json={
        "contents": [
            {
                "role": "user",
                "parts": [
                    {"text": "一张未来感 AI 工作台场景，冷色氛围，电影级布光"}
                ],
            }
        ],
        "generationConfig": {
            "responseModalities": ["TEXT", "IMAGE"],
            "temperature": 1.0,
            "topP": 0.95,
            "maxOutputTokens": 8192,
            "imageConfig": {
                "aspectRatio": "16:9",
                "imageSize": "2K",
            },
        },
    },
    timeout=60,
)
print(resp.json())

const response = await fetch(
  "https://www.geeknow.top/v1beta/models/gemini-3-pro-image-preview:generateContent",
  {
    method: "POST",
    headers: {
      Authorization: "Bearer YOUR_API_KEY",
      Accept: "application/json",
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      contents: [
        {
          role: "user",
          parts: [{ text: "一张未来感 AI 工作台场景，冷色氛围，电影级布光" }],
        },
      ],
      generationConfig: {
        responseModalities: ["TEXT", "IMAGE"],
        temperature: 1.0,
        topP: 0.95,
        maxOutputTokens: 8192,
        imageConfig: {
          aspectRatio: "16:9",
          imageSize: "2K",
        },
      },
    }),
  }
);

console.log(await response.json());

图生图示例

{
  "contents": [
    {
      "role": "user",
      "parts": [
        { "text": "融合参考图风格，输出一张高清横版海报" },
        {
          "inlineData": {
            "mimeType": "image/jpeg",
            "data": "BASE64_IMAGE_1"
          }
        },
        {
          "inlineData": {
            "mimeType": "image/jpeg",
            "data": "BASE64_IMAGE_2"
          }
        }
      ]
    }
  ],
  "generationConfig": {
    "responseModalities": ["TEXT", "IMAGE"],
    "imageConfig": {
      "aspectRatio": "16:9",
      "imageSize": "2K"
    }
  }
}

图生图字段解释

字段	图生图里的作用
`parts[].text`	告诉模型“你要怎么改、保留什么、输出成什么风格”
`parts[].inlineData.mimeType`	声明参考图格式
`parts[].inlineData.data`	参考图的 Base64 内容
`imageConfig.aspectRatio`	约束最终出图比例
`imageConfig.imageSize`	约束最终出图清晰度

参考图数量

多参考图可以按顺序追加多个 inlineData：

支持 1 张参考图
也支持多张参考图
多张时就是连续放多个 inlineData

返回值说明

Gemini 生图返回的图片通常在：

candidates[0].content.parts[].inlineData.data

这个字段可能有两种情况：

直接是 Base64 图片数据
直接是图片 URL

因此文档示例中用 BASE64_OR_URL 表示这两种可能的返回形式。

响应示例

{
  "candidates": [
    {
      "content": {
        "role": "model",
        "parts": [
          {
            "inlineData": {
              "mimeType": "image/png",
              "data": "BASE64_OR_URL"
            }
          }
        ]
      },
      "finishReason": "STOP"
    }
  ],
  "usageMetadata": {
    "promptTokenCount": 123,
    "candidatesTokenCount": 456,
    "totalTokenCount": 579,
    "trafficType": "ON_DEMAND"
  },
  "modelVersion": "gemini-3-pro-image-preview",
  "createTime": "2026-05-21T00:00:00Z"
}

Response

candidates[].content.parts[].inlineData.mimeType

string

返回图片 MIME 类型。

candidates[].content.parts[].inlineData.data

string

图片数据。可能是 Base64，也可能直接是图片 URL。

usageMetadata

object

Token 统计信息。

异步调用

Gemini 图像模型也支持通过统一图像异步入口提交任务：

POST /v1/images/generations/async

异步入口使用 OpenAI Images 兼容 JSON，而不是本页的 Gemini 原生 contents[].parts[] 请求体。model 可以传 gemini-3-pro-image-preview、gemini-2.5-flash-image-preview、gemini-3.1-flash-image-preview 或 gemini-3.1-flash-lite-image。提交成功后读取返回的 id / task_id，再通过 GET /v1/images/generations/async/{taskId} 查询状态和结果。

概览

文本系列

图像系列

视频系列

上传管理

任务管理

账户管理

Gemini 图像生成 API

Gemini 图像生成 API

支持模型

模型差异

Body

图生图请求结构

文生图 vs 图生图

图生图字段构造

图生图示例

方法与路径

请求示例

图生图示例

图生图字段解释

参考图数量

返回值说明

响应示例

Response

异步调用

相关接口

​Gemini 图像生成 API

​支持模型

​模型差异

​Body

​图生图请求结构

​文生图 vs 图生图

​图生图字段构造

​图生图示例

​方法与路径

​请求示例

​图生图示例

​图生图字段解释

​参考图数量

​返回值说明

​响应示例

​Response

​异步调用

​相关接口

Gemini 图像生成 API

支持模型

模型差异

Body

图生图请求结构

文生图 vs 图生图

图生图字段构造

图生图示例

方法与路径

请求示例

图生图示例

图生图字段解释

参考图数量

返回值说明

响应示例

Response

异步调用

相关接口