1. 模型-图片
AiApi114
  • 模型-视频
    • 可灵 图生视频
      POST
    • 可灵 图生视频任务查询
      GET
    • 可灵 文生视频
      POST
    • 可灵 文生视频任务查询
      GET
    • 生成视频
      POST
    • 查询视频任务
      GET
  • 模型-音乐(Suno)
    • 批量查询 Suno 任务
      POST
    • 查询单个 Suno 任务
      GET
    • 提交 Suno 音乐任务
      POST
  • 模型-音频
    • 文本转语音(TTS)
      POST
    • 语音转文本(转写)
      POST
    • 语音翻译为英文文本
      POST
  • 模型-对话
    • 创建对话补全(OpenAI 兼容)
      POST
    • 创建消息(Claude Messages 兼容)
      POST
  • 模型-向量
    • 创建向量嵌入
      POST
  • 模型-图片
    • 编辑图片(图生图)
      POST
    • 生成图片(文生图)
      POST
    • Gemini 图片生成(文生图/图生图)
      POST
  • 模型-重排
    • 文档重排
      POST
  1. 模型-图片

Gemini 图片生成(文生图/图生图)

POST
/v1beta/models/{model}:generateContent
通过 Gemini 对话接口生成图片。支持以下图片输入方式:

输入方式#

方式parts[] 字段说明
纯文本 → 文生图text仅传文本 prompt,根据描述生成图片
inlineData → 图生图inlineData参考图以 base64 data URI 内联传入
fileData → 图生图fileData参考图以外部 URL 传入(fileUri)

请求结构#

{
"contents": [
{
"parts": [
{"text": "图片描述 prompt"},
{"inlineData": {"mimeType": "image/jpeg", "data": "<base64>"}}
]
}
],
"generationConfig": {
"candidateCount": 1,
"seed": 42,
"imageConfig": {"aspectRatio": "16:9", "imageSize": "2K"}
}
}

参数说明#

字段类型必填说明
contents[]array是对话内容数组
contents[].rolestring否角色: user(默认)
contents[].parts[]array是消息部件
parts[].textstring建议必填文本 prompt,描述期望的图片内容
parts[].inlineDataobject否内联图片数据。字段: mimeType(图片 MIME)+ data(base64)
parts[].fileDataobject否外部图片引用。字段: mimeType + fileUri(图片 URL)
generationConfigobject否生成配置
generationConfig.candidateCountinteger否生成结果数,仅支持 1
generationConfig.seedinteger否随机种子,用于可复现生成
generationConfig.imageConfigobject否图片输出配置(亦支持 snake_case image_config)
generationConfig.imageConfig.aspectRatiostring否宽高比,如 1:1、16:9、9:16、3:2
generationConfig.imageConfig.imageSizestring否分辨率档位 512/1K/2K/4K

请求示例 1:文生图(纯文本)#

{
"contents": [{
"parts": [{"text": "a cute cat, digital art"}]
}]
}

请求示例 2:图生图(inlineData)#

{
"contents": [{
"parts": [
{"text": "watercolor style, cinematic lighting, make this scene look like an oil painting"},
{"inlineData": {"mimeType": "image/jpeg", "data": "/9j/4AAQ..."}}
]
}]
}

请求示例 3:图生图(fileData 外部 URL)#

{
"contents": [{
"parts": [
{"text": "make this a cinematic painting"},
{"fileData": {"mimeType": "image/png", "fileUri": "https://example.com/image.png"}}
]
}]
}

请求示例 4:指定宽高比与分辨率#

{
"contents": [{"parts": [{"text": "a wide landscape of mountains at sunset"}]}],
"generationConfig": {"imageConfig": {"aspectRatio": "16:9", "imageSize": "2K"}}
}

响应示例#

{
"candidates": [{
"content": {
"role": "model",
"parts": [
{"inlineData": {"mimeType": "image/png", "data": "/9j/4AAQ..."}}
]
},
"finishReason": "STOP",
"safetyRatings": [
{"category": "HARM_CATEGORY_HARASSMENT", "probability": "NEGLIGIBLE"},
{"category": "HARM_CATEGORY_HATE_SPEECH", "probability": "NEGLIGIBLE"}
]
}],
"usageMetadata": {
"promptTokenCount": 10,
"candidatesTokenCount": 100,
"totalTokenCount": 110
}
}

适用模型#

Gemini 图片模型命名模式:gemini-*-image-preview / gemini-*-pro-image-preview
图片限制:最多 7 张输入图片
不支持:candidateCount > 1

错误响应#

HTTP 状态Code说明
400invalid_requestprompt 缺失、输入图片超 7 张、candidateCount > 1
502vidu_image_errorVidu 上游返回错误码
504vidu_image_timeout异步任务轮询超时(10 分钟)

请求参数

Authorization
API Key
在 header 添加参数
Authorization
示例:
Authorization: ********************
or
Path 参数

Header 参数

Body 参数application/json必填

示例

返回响应

🟢200
application/json
图片生成结果(inlineData)
Bodyapplication/json

🟠400
🟠401
🔴502
🔴504
请求示例请求示例
Shell
JavaScript
Java
Swift
curl --location 'https://aiapi114.com/v1beta/models/:generateContent' \
--header 'Authorization: <api-key>' \
--header 'Content-Type: application/json' \
--data '{
    "cachedContent": "string",
    "contents": [
        {
            "parts": [
                {
                    "codeExecutionResult": {
                        "outcome": "string",
                        "output": "string"
                    },
                    "executableCode": {
                        "code": "string",
                        "language": "string"
                    },
                    "fileData": {
                        "fileUri": "string",
                        "mimeType": "string"
                    },
                    "functionCall": {
                        "args": null,
                        "name": "string"
                    },
                    "functionResponse": {
                        "id": [
                            0
                        ],
                        "name": "string",
                        "parts": [
                            0
                        ],
                        "response": {
                            "property1": "string",
                            "property2": "string"
                        },
                        "scheduling": [
                            0
                        ],
                        "willContinue": [
                            0
                        ]
                    },
                    "inlineData": {
                        "data": "string",
                        "mimeType": "string"
                    },
                    "mediaResolution": [
                        0
                    ],
                    "text": "string",
                    "thought": true,
                    "thoughtSignature": [
                        0
                    ],
                    "videoMetadata": [
                        0
                    ]
                }
            ],
            "role": "string"
        }
    ],
    "generationConfig": {
        "candidateCount": 0,
        "enableEnhancedCivicAnswers": true,
        "frequencyPenalty": 0,
        "imageConfig": [
            0
        ],
        "logprobs": 0,
        "maxOutputTokens": 0,
        "mediaResolution": "string",
        "presencePenalty": 0,
        "responseJsonSchema": [
            0
        ],
        "responseLogprobs": true,
        "responseMimeType": "string",
        "responseModalities": [
            "string"
        ],
        "responseSchema": null,
        "seed": 0,
        "speechConfig": [
            0
        ],
        "stopSequences": [
            "string"
        ],
        "temperature": 0,
        "thinkingConfig": {
            "includeThoughts": true,
            "thinkingBudget": 0,
            "thinkingLevel": "string"
        },
        "topK": 0,
        "topP": 0
    },
    "requests": [
        {}
    ],
    "safetySettings": [
        {
            "category": "string",
            "threshold": "string"
        }
    ],
    "systemInstruction": {
        "parts": [
            {
                "codeExecutionResult": {
                    "outcome": "string",
                    "output": "string"
                },
                "executableCode": {
                    "code": "string",
                    "language": "string"
                },
                "fileData": {
                    "fileUri": "string",
                    "mimeType": "string"
                },
                "functionCall": {
                    "args": null,
                    "name": "string"
                },
                "functionResponse": {
                    "id": [
                        0
                    ],
                    "name": "string",
                    "parts": [
                        0
                    ],
                    "response": {
                        "property1": "string",
                        "property2": "string"
                    },
                    "scheduling": [
                        0
                    ],
                    "willContinue": [
                        0
                    ]
                },
                "inlineData": {
                    "data": "string",
                    "mimeType": "string"
                },
                "mediaResolution": [
                    0
                ],
                "text": "string",
                "thought": true,
                "thoughtSignature": [
                    0
                ],
                "videoMetadata": [
                    0
                ]
            }
        ],
        "role": "string"
    },
    "toolConfig": {
        "functionCallingConfig": {
            "allowedFunctionNames": [
                "string"
            ],
            "mode": "string"
        },
        "includeServerSideToolInvocations": true,
        "retrievalConfig": {
            "languageCode": "string",
            "latLng": {
                "latitude": 0,
                "longitude": 0
            }
        }
    },
    "tools": [
        0
    ]
}'
响应示例响应示例
200 - 示例 1
{
    "candidates": [
        {
            "content": {
                "parts": [
                    {
                        "codeExecutionResult": {
                            "outcome": "string",
                            "output": "string"
                        },
                        "executableCode": {
                            "code": "string",
                            "language": "string"
                        },
                        "fileData": {
                            "fileUri": "string",
                            "mimeType": "string"
                        },
                        "functionCall": {
                            "args": null,
                            "name": "string"
                        },
                        "functionResponse": {
                            "id": [
                                0
                            ],
                            "name": "string",
                            "parts": [
                                0
                            ],
                            "response": {
                                "property1": "string",
                                "property2": "string"
                            },
                            "scheduling": [
                                0
                            ],
                            "willContinue": [
                                0
                            ]
                        },
                        "inlineData": {
                            "data": "string",
                            "mimeType": "string"
                        },
                        "mediaResolution": [
                            0
                        ],
                        "text": "string",
                        "thought": true,
                        "thoughtSignature": [
                            0
                        ],
                        "videoMetadata": [
                            0
                        ]
                    }
                ],
                "role": "string"
            },
            "finishReason": "string",
            "index": 0,
            "safetyRatings": [
                {
                    "category": "string",
                    "probability": "string"
                }
            ]
        }
    ],
    "promptFeedback": {
        "blockReason": "string",
        "safetyRatings": [
            {
                "category": "string",
                "probability": "string"
            }
        ]
    },
    "usageMetadata": {
        "cachedContentTokenCount": 0,
        "candidatesTokenCount": 0,
        "candidatesTokensDetails": [
            {
                "modality": "string",
                "tokenCount": 0
            }
        ],
        "promptTokenCount": 0,
        "promptTokensDetails": [
            {
                "modality": "string",
                "tokenCount": 0
            }
        ],
        "thoughtsTokenCount": 0,
        "toolUsePromptTokenCount": 0,
        "toolUsePromptTokensDetails": [
            {
                "modality": "string",
                "tokenCount": 0
            }
        ],
        "totalTokenCount": 0
    }
}
上一页
生成图片(文生图)
下一页
文档重排
Built with