文本生成-流

开发中

POST

/v1beta/models/gemini-3-pro-preview:streamGenerateContent

请替换接口中 {model-name} 的模型名字。
官方文档：https://ai.google.dev/gemini-api/docs/text-generation?hl=zh-cn

请求参数

Header 参数

Body 参数application/json

示例

{
    "systemInstruction": {
        "parts": [
            {
                "text": "You are a cat. Your name is Neko."
            }
        ]
    },
    "contents": [
        {
            "role": "user",
            "parts": [
                {
                    "text": "Hello there"
                }
            ]
        }
    ],
    "generationConfig": {
        "temperature": 1,
        "topP": 1,
        "thinkingConfig": {
            "includeThoughts": true,
            "thinkingBudget": 26240
        }
    }
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location 'https://api.modelport.ai/v1beta/models/gemini-3-pro-preview:streamGenerateContent' \
--header 'Accept: application/json' \
--header 'Authorization: Bearer <您的真实API_KEY>' \
--header 'Content-Type: application/json' \
--data '{
    "systemInstruction": {
        "parts": [
            {
                "text": "You are a cat. Your name is Neko."
            }
        ]
    },
    "contents": [
        {
            "role": "user",
            "parts": [
                {
                    "text": "Hello there"
                }
            ]
        }
    ],
    "generationConfig": {
        "temperature": 1,
        "topP": 1,
        "thinkingConfig": {
            "includeThoughts": true,
            "thinkingBudget": 26240
        }
    }
}'

返回响应

🟢200成功

application/json

Bodyapplication/json

示例

{
    "candidates": [
        {
            "content": {
                "role": "model",
                "parts": [
                    {
                        "text": "<think>**Formulating a Response**\n\nI'm currently crafting a friendly and helpful response in Chinese to the user's greeting. I plan to offer assistance and be prepared to switch to English if the user prefers, since that's my native language. I want to ensure the interaction is comfortable and productive.\n\n\n</think>\n\n你好！我能为你做些什么？"
                    }
                ]
            },
            "finishReason": "STOP",
            "index": 0,
            "safetyRatings": []
        }
    ],
    "usageMetadata": {
        "promptTokenCount": 8,
        "candidatesTokenCount": 78,
        "totalTokenCount": 86,
        "thoughtsTokenCount": 0,
        "promptTokensDetails": "None"
    }
}

修改于 2026-04-20 00:49:03

聊天

Nano Banana