图像识别

POST

{{url}}/v1beta/models/gemini-2.5-flash:generateContent

最后修改时间：2025-11-29 15:42:08

请求参数

Query 参数

Header 参数

Body 参数application/json

示例

{
    "contents": [
        {
            "parts": [
                {
                    "inline_data": {
                        "mime_type": "image/jpeg",
                        "data": "输入Base64"
                    }
                },
                {
                    "text": "Caption this image."
                }
            ]
        }
    ]
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location -g --request POST 'http://api.example.com/v1beta/models/gemini-2.5-flash:generateContent?key=sk-xxx' \
--header 'Content-Type: application/json' \
--data-raw '{
    "contents": [
        {
            "parts": [
                {
                    "inline_data": {
                        "mime_type": "image/jpeg",
                        "data": "输入Base64"
                    }
                },
                {
                    "text": "Caption this image."
                }
            ]
        }
    ]
}'

返回响应

🟢200成功

application/json

Body

示例

{
    "candidates": [
        {
            "content": {
                "role": "model",
                "parts": [
                    {
                        "text": "Here are a few caption options for the image:\n\n1.  An aerial view reveals a rural landscape dominated by agricultural fields, some showing contour farming patterns, and several prominent wind turbines.\n2.  Agricultural fields, roads, and a small village are visible in this aerial image, alongside multiple wind turbines harnessing renewable energy in the rural environment.\n3.  A satellite image displays a mix of green crops and tilled brown earth forming striped patterns across undulating rural terrain, with several wind turbines generating power.\n4.  Wind turbines stand amidst a patchwork of agricultural fields and rolling hills in a rural setting."
                    }
                ]
            },
            "finishReason": "STOP",
            "avgLogprobs": -3.5620666503906251
        }
    ],
    "usageMetadata": {
        "promptTokenCount": 1294,
        "candidatesTokenCount": 125,
        "totalTokenCount": 2490,
        "trafficType": "ON_DEMAND",
        "promptTokensDetails": [
            {
                "modality": "TEXT",
                "tokenCount": 4
            },
            {
                "modality": "IMAGE",
                "tokenCount": 1290
            }
        ],
        "candidatesTokensDetails": [
            {
                "modality": "TEXT",
                "tokenCount": 125
            }
        ],
        "thoughtsTokenCount": 1071
    },
    "modelVersion": "gemini-2.5-flash",
    "createTime": "2025-11-29T15:41:23.445792Z",
    "responseId": "IxQraeCaG_O24_UP9PW30QQ"
}

修改于 2025-11-29 15:42:08

PDF识别

视频识别