跳到内容

Nano Banana ​

更新时间

Endpoint: POST https://vod.intl.tencentcloudapi.comAction: CreateAigcImageTask

Nano Banana 是图片生成模型,支持文生图与多图参考生成 / 编辑(3.6 / 3.1 最多可传入 14 张参考图)。输出分辨率 1K~4K;最新版本 3.6(Nano Banana 2.1)新增 1:8 / 8:1 等超宽比例,3.1-lite 适合追求速度的场景。

基本信息 ​

项目值
ModelNameGEM
ModelVersion3.6 / 3.1 / 3.1-lite / 3.0 / 2.5
默认值ModelVersion=3.1 / AspectRatio=1:1
护栏解除支持不支持

各版本支持规格 ​

版本对应模型最大参考图分辨率宽高比
3.6Nano Banana 2.114 张(其中角色图最多 4 张)1K / 2K / 4K1:1 / 1:4 / 1:8 / 2:3 / 3:2 / 3:4 / 4:1 / 4:3 / 4:5 / 5:4 / 8:1 / 9:16 / 16:9 / 21:9
3.1Nano Banana 214 张1K / 2K / 4K同 3.6
3.1-liteNano Banana 2 Lite14 张1K / 2K / 4K同 3.6
3.0Nano Banana Pro14 张1K / 2K / 4K1:1 / 2:3 / 3:2 / 3:4 / 4:1 / 4:3 / 4:5 / 5:4 / 9:16 / 16:9 / 21:9
2.5Nano Banana3 张1K / 2K / 4K同 3.0

所有版本均支持文生图、参考生图与图片编辑。

请求参数 ​

参数必填类型说明
ModelName必填String固定值 GEM
ModelVersion可选String3.6 / 3.1 / 3.1-lite / 3.0 / 2.5
Prompt必填String生成提示词
FileInfos.N可选Array参考图片(image-to-image)。Category 为 Image。张数上限见上表
OutputConfig.Resolution可选String1K / 2K / 4K
OutputConfig.AspectRatio可选String1:1 / 16:9 / 9:16 / 4:3 / 3:4 等,各版本可选值见上表
OutputConfig.OutputImageCount可选Integer生成张数(各模型上限不同)
ExtInfo可选String扩展参数。可通过 thinkingLevel 自定义思考等级:minimal / medium / high,默认 medium

请求示例 ​

3.6 / 文生图 ​

json
{
  "SubAppId": 123456789,
  "ModelName": "GEM",
  "ModelVersion": "3.6",
  "Prompt": "an artist in a bunny costume copies the image on her monitor onto a canvas as an oil painting; the photo itself captures her painting this recursive image",
  "OutputConfig": {
    "StorageMode": "Temporary",
    "Resolution": "1K",
    "AspectRatio": "9:16"
  },
  "SessionContext": "gem36-t2i-001"
}

3.6 / 多图融合 ​

json
{
  "SubAppId": 123456789,
  "ModelName": "GEM",
  "ModelVersion": "3.6",
  "FileInfos": [
    {
      "Type": "Url",
      "Category": "Image",
      "Url": "https://<host>/image-1.png"
    },
    {
      "Type": "Url",
      "Category": "Image",
      "Url": "https://<host>/image-2.png"
    }
  ],
  "Prompt": "Blend image 1 and image 2 into a single coherent scene.",
  "OutputConfig": {
    "StorageMode": "Temporary",
    "Resolution": "2K",
    "AspectRatio": "16:9"
  },
  "SessionContext": "gem36-i2i-001"
}

3.1 / 文生图 ​

json
{
  "SubAppId": 123456789,
  "ModelName": "GEM",
  "ModelVersion": "3.1",
  "Prompt": "a serene japanese garden, soft morning light",
  "OutputConfig": {
    "AspectRatio": "1:1",
    "OutputImageCount": 1,
    "StorageMode": "Temporary"
  },
  "SessionContext": "job-001"
}

参考图片请求 ​

json
{
  "SubAppId": 123456789,
  "ModelName": "GEM",
  "ModelVersion": "3.1",
  "Prompt": "Change the background to a quiet garden; preserve the subject.",
  "OutputConfig": {
    "AspectRatio": "1:1",
    "OutputImageCount": 1,
    "StorageMode": "Temporary"
  },
  "SessionContext": "job-001",
  "FileInfos": [
    {
      "Type": "Url",
      "Category": "Image",
      "Url": "https://<host>/reference.jpg"
    }
  ]
}

响应示例 ​

json
{
  "AigcImageTask": {
    "Status": "FINISH",
    "ErrCode": 0,
    "Progress": 100,
    "Output": {
      "FileInfos": [
        {
          "FileUrl": "http://<host>.vod2.myqcloud.com/.../aigcImageGenFile.png",
          "ExpireTime": "2026-08-01T10:29:48Z",
          "MetaData": {
            "Width": 1024,
            "Height": 1024,
            "Container": "png"
          }
        }
      ]
    }
  }
}

特殊设置 ​

附加选项 ​

Google Gemini Image 系列

支持文本→图片生成与参考图片生成(i2i)。

自定义思考等级(thinkingLevel) ​

通过 ExtInfo.AdditionalParameters.thinkingLevel 调整模型生成前的推理深度,可选 minimal / medium / high,默认 medium。等级越高,构图与指令遵循越细致,耗时也越长。

json
{
  "SubAppId": 123456789,
  "ModelName": "GEM",
  "ModelVersion": "3.6",
  "FileInfos": [
    {
      "Type": "Url",
      "Category": "Image",
      "Url": "https://<host>/reference.png"
    }
  ],
  "Prompt": "Turn this sketch into a detailed 16:9 storyboard with 8 panels.",
  "OutputConfig": {
    "StorageMode": "Temporary"
  },
  "ExtInfo": "{\"AdditionalParameters\": \"{\\\"thinkingLevel\\\":\\\"high\\\"}\"}",
  "SessionContext": "gem36-think-001"
}

ModelName 写 GEM 或 GG 均可

接入指南中 Nano Banana 系列的 ModelName 记作 GEM/GG,两者等价;本页示例统一使用 GEM。

基于 VitePress 构建 · 部署于腾讯云 EdgeOne Pages