1. Google
toiotech
  • toiotech接口说明
  • 已支持的模型列表
  • 大模型官方文档
  • Claude thinking 多轮对话
  • thinking相关说明
  • 已支持的模型列表
    • 语言大模型
      • OpenAI
        • gpt-4
        • gpt-4.1
        • gpt-4o
        • gpt-4o-mini
        • gpt-5
        • gpt-5-chat
        • gpt-5-mini
        • gpt-5-nano
        • gpt-5.1
        • gpt-5.1-chat-latest
        • gpt-5.1-codex(仅支持/v1/responses)
        • gpt-5.2
        • gpt-5.2(responses)
        • gpt-5.2-chat
        • gpt-5.2-pro(仅支持/v1/responses)
        • gpt-oss-120b
        • o1
        • o3
        • o3-mini
        • o3-deep-research
        • o3-pro
        • o4-mini
        • o4-mini-deep-research
        • gpt-5.4
        • gpt-5.4-pro
      • Google
        • gemini-2.0-flash
          POST
        • gemini-2.5-computer-use-preview-10-2025
          POST
        • gemini-2.5-flash
          POST
        • gemini-2.5-flash-lite
          POST
        • gemini-2.5-flash-lite-preview
          POST
        • gemini-2.5-pro
          POST
        • gemini-3-flash-preview
          POST
        • gemini-3-pro-preview
          POST
        • gemini-robotics-er-1.5-preview
          POST
        • gemini-(thinking_config)openai请求格式
          POST
        • gemini-3.1-pro-preview(多模态)
          POST
        • gemini-3.1-flash-lite-preview(多模态)
          POST
        • gemini-3.1-flash-lite-preview(thinking)
          POST
        • gemini-2.5-pro(原生请求格式)
          POST
      • Anthropic
        • 通用接口
        • claude-opus-4-5-20251101
        • claude-opus-4-20250514
        • claude-sonnet-4-20250514
        • claude-sonnet-4-5-20250929
        • claude-(Interleaved thinking)
        • claude-opus-4-6
        • claude-sonnet-4-6
      • DeepSeek
        • deepseek-chat
        • deepseek-reasoner
      • MiniMax
        • abab6.5s-chat
        • minimax-m2
        • minimax-m2.5
      • Moonshot
        • kimi-k2
        • kimi-k2-0711-preview
        • kimi-k2-0905
        • kimi-k2-instruct
        • kimi-k2-thinking
        • moonshot-v1-128k
        • moonshot-v1-32k
        • moonshot-v1-8k
        • moonshot-v1-128k-vision-preview
        • moonshot-v1-32k-vision-preview
        • moonshot-v1-8k-vision-preview
        • kimi-k2.5
        • 获取系统工具
        • 执行系统工具
      • xAI
        • grok-4
        • grok-4-1-fast-non-reasoning
        • grok-4-1-fast-reasoning
      • 字节跳动
        • 通用接口
        • doubao-1.8图片理解
        • doubao-1.8(视频理解)
        • doubao-seed-2-0-pro-260215(图片理解)
        • doubao-seed-2-0-pro-260215(视频理解)
      • 智谱
        • glm-4-plus
        • glm-4.5
        • glm-4.6
        • glm-4.7
        • glm-5
      • 阿里巴巴
        • qwen-max
        • qwen-max-2025-01-25
        • qwen-max-latest
        • qwen-plus
        • qwen-plus-latest
        • qwen-turbo
        • qwen2.5-7b-instruct
        • qwen2.5-32b-instruct
        • qwen2.5-72b-instruct
        • qwen3-coder-plus
        • qwen3-coder-plus-2025-07-22
        • qwen3-max
        • qwen3-max-preview
        • qwen3-235b-a22b
        • qwen3-235b-a22b-instruct-2507
        • qwen3-235b-a22b-thinking-2507
        • qwen3-8b
        • qwen3-coder-30b-a3b-instruct
        • qwen3-next-80b-a3b-instruct
        • qwen3-next-80b-a3b-thinking
        • qwen-vl-max(视觉理解)
        • qwen-vl-plus(视觉理解)
        • qwen3-vl-235b-a22b-instruct(视觉理解)
        • qwen3-vl-235b-a22b-thinking(视觉理解)
        • qwen3-vl-30b-a3b-thinking(图片理解)
        • qwen3-vl-30b-a3b-thinking(视频理解)
        • qwen3-max-2026-01-23(qwen3-max-thinking)
        • qwen3-max-2026-01-23(非思考模式)
        • qwen3.5-397b-a17b(多模态)
        • qwen3.5-plus(多模态)
        • qwen3.5-flash(多模态)
        • qwen3.5-122b-a10b(多模态)
        • qwen3.5-35b-a3b
        • qwen3.5-omni-plus(视频+文本)
        • qwen3.5-omni-plus(音频+文本)
        • qwen3.5-omni-plus(图片+文本)
        • qwen3.5-omni-flash(图片+文本)
      • 百度
        • ernie-x1-32k
        • ernie-5.0-thinking-exp
        • ernie-5.0-thinking-preview
      • 腾讯
        • hunyuan-t1-latest
      • 讯飞
        • SparkDesk-v4.0
      • 蚂蚁
        • Ling-1T
        • Ring-1T
      • 阶跃星辰
        • step-3
      • 小米
        • mimo-v2-pro
        • mimo-v2-omni(多模态理解)
    • 图片生成
      • 通用接口
        • 图片生成
        • 图片编辑
      • OpenAI
        • gpt-image系列图片生成
        • gpt-image系列图片编辑
        • dall-e-3图片生成
      • Google
        • gemini-3-pro-image-preview
        • gemini-3.1-flash-image-preview
        • gemini-2.5-flash-image
        • gemini-2.5-flash-image-preview
      • Anthropic
      • BFL
        • flux-1-kontext-pro
      • ImagineArt
        • imagineart-1.5-preview
      • 字节跳动豆包
        • doubao-seedream-5-0-260128(文生图)
        • doubao-seedream-5-0-260128(图生图)
    • 音视频处理
      • OpenAI
        • gpt-4o-audio-preview
        • gpt-audio
      • 阿里巴巴
        • qwen3-tts-flash
      • Elevenlabs
        • eleven_v3
      • MiniMax
        • 通用接口
    • 视频生成
      • Vidu
        • vidu-2.0-img2video
        • viduq2-pro提交图生视频任务
        • viduq3-pro图生视频
        • viduq3-pro文生视频
      • 字节跳动
        • 提交图生视频生成任务
        • 提交文生视频生成任务
      • 快手
        • kling提交图生视频任务
        • kling提交文生视频任务
        • kling-v3提交图生视频任务
        • kling-v3提交文生视频任务
      • 腾讯
        • hunyuan-video-v1.5
      • 海螺
        • 提交图生视频任务
        • 提交文生视频任务
      • sora
        • sora-2
        • sora-2-pro
      • Google
        • veo-3.1-generate-preview
        • veo-3.1-fast-generate-preview
      • 阿里巴巴
        • wan2.6-i2v
        • wan2.6-t2v
      • pixverse
        • pixverse-v5.5-t2v
        • pixverse-v5.5-i2v
        • pixverse-v5.6-i2v
        • pixverse-v5.6-t2v
      • ltx
        • 文生视频
        • 图生视频
      • worldlabs
        • 文生视频(world)
        • 图生视频(world)
        • 多图生视频(world)
      • runawayml
        • 提交gen4.5(文生视频)
        • 提交gen4.5(图生视频)
    • 多模态模型
      • 智谱
        • glm-4.5v
        • glm-4.6v
      • 阿里巴巴
        • qwen2.5-omni-7b
        • qwen2.5-vl-72b-instruct
        • qwen3-omni-flash-2025-12-01
      • 日日新
        • SenseNova-V6-Pro
    • Realtime
      • OpenAI
        • gpt-4o-mini-realtime-preview
        • gpt-4o-realtime-preview
        • gpt-realtime-1.5
      • qwen阿里云
        • qwen3-omni-flash-realtime
      • gemini
        • gemini-3.1-flash-live-preview
    • 向量化
      • OpenAI
        • text-embedding-3-large
        • text-embedding-3-small
        • text-embedding-ada-002
        • text-embedding-v1
    • 代码解释器
      • qwen3-模型
  • 已支持的搜索服务
    • serper
      • get方式
      • post方式
  • gemini
    • openai格式
      • gpt-5.2
      • openai格式
      • gemini-开启thinking
    • google原生格式
      • gemini-robotics-er-1.5-preview
      • google格式
  • anthropic
    • anthropic格式
      • anthropic格式
      • tools调用
      • thinking
      • image理解
      • code_execution
      • 关闭thinking
    • openai格式
      • openai格式
  • openai
    • chat/completions
    • 流式/chat/completions
    • responses
  • deepseek
    • chat/completions
  • qwen
    • qwen提交文生视频任务
    • qwen提交图生视频任务
    • qwen3-omni-flash
    • chat/completions(开启thinking)
    • chat/completions
  • 获取异步任务结果
    GET
  1. Google

gemini-(thinking_config)openai请求格式

POST
/v1/chat/completions
https://ai.google.dev/gemini-api/docs/thinking?hl=zh-cn
官方文档

Request

Authorization
Provide your bearer token in the
Authorization
header when making requests to protected resources.
Example:
Authorization: Bearer ********************
Body Params application/jsonRequired

Example
{
    "messages": [
        {
            "role": "user",
            "content": "你好呀?,8.11和8.9哪个大?为什么?"
        }
    ],
    "model": "gemini-3-pro-preview",
    "temperature": 0.1,
    "top_p": 1.0,
    "stream": false,
    "extra_body": {
        "google": {
            "thinking_config": {
                "include_thoughts": true,
                "thinking_budget": 1024
            }
        }
    }
}

Request Code Samples

Shell
JavaScript
Java
Swift
Go
PHP
Python
HTTP
C
C#
Objective-C
Ruby
OCaml
Dart
R
Request Request Example
Shell
JavaScript
Java
Swift
curl --location --request POST 'https://api.toiotech.com/v1/chat/completions' \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data-raw '{
    "messages": [
        {
            "role": "user",
            "content": "你好呀?,8.11和8.9哪个大?为什么?"
        }
    ],
    "model": "gemini-3-pro-preview",
    "temperature": 0.1,
    "top_p": 1.0,
    "stream": false,
    "extra_body": {
        "google": {
            "thinking_config": {
                "include_thoughts": true,
                "thinking_budget": 1024
            }
        }
    }
}'

Responses

🟢200OK
application/json
Body

Example
{
    "id": "chatcmpl-20260115181125604771578gH26U8AF",
    "model": "gemini-3-pro-preview",
    "object": "chat.completion",
    "created": 1768471904,
    "choices": [
        {
            "index": 0,
            "message": {
                "role": "assistant",
                "content": "你好呀!👋\n\n**8.9 比 8.11 大。**\n\n这里有两个简单的理由来解释为什么:\n\n**方法一:按数位比较(最标准的方法)**\n比较小数大小时,我们要从左往右一位一位地比:\n1.  先看**整数部分**:两个数都是 **8**,打平手。\n2.  再看**小数点后第一位(十分位)**:\n    *   8.**9** 的十分位是 **9**\n    *   8.**1**1 的十分位是 **1**\n3.  因为 **9 比 1 大**,所以不用看后面的数字了,**8.9 更大**。\n\n**方法二:补零法(最直观的方法)**\n为了方便比较,我们可以把它们的小数位数变得一样多(在末尾补0不改变大小):\n*   8.9 可以写成 **8.90**\n*   8.11 还是 **8.11**\n\n这时候你就看得很清楚了:**8.90 比 8.11 大**(就像 90 分比 11 分多一样)。\n\n**为什么容易弄错?**\n很多人会直觉地觉得“11 比 9 大”,所以认为 8.11 大。这是把小数当成整数来比了。但在小数的世界里,靠前的位数(十分位)权重更大,**0.9 (9个0.1) 远比 0.11 (1个0.1加1个0.01) 要大得多**。",
                "reasoning_content": "**Internal Dialogue: Breaking Down the Decimal Comparison**\n\nOkay, here's how I'm approaching this. The user starts with a friendly \"Hello,\" and the core of the question is a decimal comparison: \"Which is bigger, 8.11 or 8.9?\" and the follow-up, \"Why?\". My goal is to give a clear and accurate explanation, and the appropriate language is Mandarin Chinese. I need to be pedagogically sound, as the \"Why?\" implies a desire for understanding, not just an answer.\n\nFirst, let's nail down the mathematical concept. This is about comparing decimal numbers, which means understanding place value and how it works in the tenths and hundredths places (and so on). I have two main ways to explain this. One is place value, the other is making the decimal lengths equal.\n\nThe direct answer is easy: 8.9 is bigger. But the *why* is crucial. I need to provide a good answer, and address the common pitfall here, which is the misconception that \"11\" is bigger than \"9\" because \"11\" is the bigger whole number.\n\nSo, I'll structure my explanation around a couple of methods:\n\n*   **Place Value Comparison:** We look at the integers (both are 8, so we proceed). Then, the tenths place. 8.9 has a 9 in the tenths place, and 8.11 has a 1. 9 is greater than 1, so 8.9 is greater. I'll make sure to use the correct terminology here, \"十分位\" etc.\n*   **\"Padding with Zeros\" Method:** We can think of 8.9 as 8.90, just like we can write 5 as 5.0. This makes the lengths of the decimal parts the same. Now we're comparing 8.90 and 8.11, and we have 90 > 11.\n\nI'll also need to explicitly address the misconception. Many people will see 8.11 and think it’s bigger because of the “11” versus the “9”. I'll have to gently point out that this is where whole number logic fails. I need to clarify that 0.11 is less than 0.9.\n\nDrafting the response in Chinese is important. I'll start with a simple greeting \"你好呀!\", then the clear conclusion: \"8.9 is larger\". Then I can carefully explain the two methods, making sure to show how they both lead to the same result. The language should be friendly but clear. I'll avoid complex terminology. I'll include the addressing of the misconception at the end to reinforce the point.\n\nFinally, I'll revisit it and look for the best phrasing and structure. Then I can make sure the response is accurate and the reasoning is sound, providing the user with a valuable explanation.\n\n\n"
            },
            "finish_reason": "stop"
        }
    ],
    "usage": {
        "prompt_tokens": 17,
        "completion_tokens": 1346,
        "total_tokens": 1363,
        "prompt_tokens_details": {
            "cached_tokens": 0,
            "text_tokens": 17,
            "audio_tokens": 0,
            "image_tokens": 0
        },
        "completion_tokens_details": {
            "text_tokens": 0,
            "audio_tokens": 0,
            "reasoning_tokens": 991
        },
        "input_tokens": 0,
        "output_tokens": 0,
        "input_tokens_details": null,
        "claude_cache_creation_5_m_tokens": 0,
        "claude_cache_creation_1_h_tokens": 0
    }
}
Modified at 2026-03-09 03:47:18
Previous
gemini-robotics-er-1.5-preview
Next
gemini-3.1-pro-preview(多模态)
Built with