AuraX API Documentation (僅供參考)

全面了解如何將 Aurax AI 無縫接入您的應用程式。我們提供與 OpenAI 完全兼容的 API 格式,讓您以極低成本調用全球頂級模型。
Learn how to seamlessly integrate Aurax AI into your applications. We provide an OpenAI-compatible API format, allowing you to access top-tier global models at a fraction of the cost.

# 快速開始指南 / Quick Start Guide

1. 獲取 Token / Get API Key

登入後台系統,進入「令牌 (Token)」頁面,點擊「添加令牌」即可獲取您的專屬 API Key。
Log into the dashboard, navigate to the "Tokens" page, and click "Add Token" to generate your exclusive API Key.

2. 接口配置 / Base URL Configuration

我們完美兼容 OpenAI 的 SDK,您只需要將原有的 Base URL 替換為以下任一地址:
We are fully compatible with OpenAI SDKs. Simply replace your existing Base URL with any of the following:

https://api.auraxai.dev
https://api.auraxai.dev/v1
https://api.auraxai.dev/v1/chat/completions

注意:不同客戶端 (如 NextChat、LobeChat) 可能需要使用不同的 Base URL 後綴,建議依次嘗試以上地址。

3. 配置範例 / Configuration Example

{
  "base_url": "https://api.auraxai.dev",
  "api_key": "your_aurax_token_here",
  "model": "gpt-4o"
}

# 發出首個請求 / Make Your First Request

您可以將下面的命令粘貼到您的終端機 (Terminal) 中以運行您的第一個 API 請求。請確保將 YOUR_API_KEY 替換為您的真實金鑰。
Paste the following command into your terminal to run your first API request. Make sure to replace YOUR_API_KEY with your actual key.

curl https://api.auraxai.dev/v1/chat/completions \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "gpt-4o-mini",
    "messages": [{"role": "user", "content": "Say this is a test!"}],
    "temperature": 0.7
  }'

# API 分組類型對比 / Routing & Billing

💡 關於官方費率與價格差異 / About Official Rates

為確保透明度,我們引入了「官方費率」概念:
• 當分組費率 = 1 時:官方價格扣 $1,我們平台也只扣 $1。
• 當分組費率 = 1.65 時:官方價格扣 $1,我們平台扣 $1.65。
不同的路由渠道成本不同,用戶可根據對穩定性及價格的需求自由選擇。

分組名稱 / Group類型說明 / Description費率倍數 / Multiplier支援模型 / Supported
default 默認混合 ChatGPT + Claude + MJ + 國產(DeepSeek+Qwen)官方費率 * 1All
純 AZ只有 ChatGPT 模型(AZ渠道) + 國產模型官方費率 * 1.5OpenAI, 國產
逆向 / Reverse支援 GPT + Claude + Gemini + Grok官方費率 * 1.4All
直連克勞德Claude (Anthropic 官方直連渠道,極致穩定)官方費率 * 16Claude

# 代理接口調用地址 / API Endpoints

🌐 推薦節點 (全域加速) / Global CDN

  • CDN 分站 (全球 60+ 節點)https://api.auraxai.dev
  • CF 站 (Cloudflare CDN)https://cf.auraxai.dev

🔍 其他服務地址 / Other Services

  • 免登錄 API 接口 (Proxy)https://api.proxy.auraxai.dev

# 線上除錯與 FAQ / Debugging & FAQ

1. 報錯:Invalid authorization header
這通常是因為您沒有在請求頭 (Headers) 中正確設置 Token。請確保您加入了 Authorization: Bearer YOUR_API_KEY,並且沒有多餘的空格。
2. 無法請求地址 api.openai.com
如果您是使用我們提供的中轉服務,請務必將代碼中的 api.openai.com 完全替換為我們的代理地址 api.auraxai.dev

數據模型 / Data Models

聊天完成對象 (Chat Completion Object)

參數 / Parameter類型 / Type描述 / Description
idstring聊天完成的唯一標識符
choicesarray聊天完成選項列表。如果 n 大於 1,可以有多個選項
createdinteger創建聊天完成的 Unix 時間戳 (秒)
modelstring用於聊天完成的模型
system_fingerprintstring該指紋表示模型運行的後端配置
objectstring對象類型,總是 `chat.completion`
usageobject完成請求的使用統計信息
{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "created": 1677652288,
  "model": "gpt-3.5-turbo-0613",
  "system_fingerprint": "fp_44709d6fcb",
  "choices": [{
    "index": 0,
    "message": {
      "role": "assistant",
      "content": "\n\nHello there, how may I assist you today?"
    },
    "finish_reason": "stop"
  }],
  "usage": {
    "prompt_tokens": 9,
    "completion_tokens": 12,
    "total_tokens": 21
  }
}

聊天完成塊對象 (Chat Completion Chunk Object)

參數 / Parameter類型 / Type描述 / Description
idstring聊天完成的唯一標識符。每個塊具有相同的 ID
choicesarray聊天完成選項列表。如果 n 大於 1,可以有多個選項
createdinteger創建聊天完成的 Unix 時間戳 (秒)。每個塊具有相同的時間戳
modelstring生成完成的模型
system_fingerprintstring該指紋表示模型運行的後端配置
objectstring對象類型,總是 `chat.completion.chunk`
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"role":"assistant","content":""},"finish_reason":null}]}
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"content":"!"},"finish_reason":null}]}
....
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"content":" today"},"finish_reason":null}]}
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{"content":"?"},"finish_reason":null}]}
{"id":"chatcmpl-123","object":"chat.completion.chunk","created":1694268190,"model":"gpt-3.5-turbo-0613", "system_fingerprint": "fp_44709d6fcb", "choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

API 參考 / API Reference

GPT-4o-audio 音頻對話

POST/v1/chat/completions

支援文本與音頻模態輸出的對話模型。 / Supports text and audio modality output.

Example Request

{
  "model": "gpt-4o-audio-preview",
  "modalities": ["text", "audio"],
  "audio": {
    "voice": "alloy",
    "format": "wav"
  },
  "messages": [
    { "role": "user", "content": "Is a golden retriever a good family dog?" }
  ]
}

音頻轉文字 / Audio Transcription (whisper-1)

POST/v1/audio/transcriptions

將音頻轉換為文字。 / Transcribe audio into text using whisper-1.

Example Request

// Content-Type: multipart/form-data
{
  "file": "<Binary Audio File>",
  "model": "whisper-1",
  "language": "zh",
  "response_format": "json"
}

音頻轉文字 / Audio Transcription (gpt-4o-transcribe)

POST/v1/audio/transcriptions

使用 GPT-4o 進行高精度音頻轉文字。 / High-accuracy transcription using GPT-4o.

Example Request

// Content-Type: multipart/form-data
{
  "file": "<Binary Audio File>",
  "model": "gpt-4o-transcribe",
  "response_format": "json"
}

創建語音 / Create Speech (gpt-4o-mini-tts)

POST/v1/audio/speech

文本轉語音功能。 / Text-to-Speech generation.

Example Request

{
  "model": "gpt-4o-mini-tts",
  "input": "The quick brown fox jumped over the lazy dog.",
  "voice": "alloy"
}

創建聊天補全 / Chat Completions (Stream)

POST/v1/chat/completions

標準流式對話輸出。 / Standard streaming chat completions.

Example Request

{
  "model": "gpt-4o-mini",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "你好 / Hello" }
  ],
  "temperature": 1,
  "stream": true,
  "stream_options": { "include_usage": true }
}

創建聊天補全 / Chat Completions (Non-stream)

POST/v1/chat/completions

非流式對話輸出。 / Non-streaming chat completions.

Example Request

{
  "model": "gpt-4o",
  "messages": [
    { "role": "user", "content": "nihao。" }
  ],
  "max_tokens": 1000
}

創建聊天識圖 / Chat Vision (Stream)

POST/v1/chat/completions

支援圖像理解的流式輸出。 / Vision capabilities with streaming.

Example Request

{
  "model": "gpt-4o",
  "messages": [
    {
      "role": "user",
      "content": [
        { "type": "text", "text": "這張圖片裡有什麼?請詳細描述。 / What's in this image?" },
        { "type": "image_url", "image_url": { "url": "https://example.com/image.png" } }
      ]
    }
  ],
  "stream": true
}

創建聊天創作圖 / Chat Image Generation

POST/v1/chat/completions

透過對話接口生成或編輯圖片。 / Generate or edit images via Chat API.

Example Request

{
  "model": "gpt-4o-image-vip",
  "messages": [
    {
      "role": "user",
      "content": [
        { "type": "text", "text": "美化一下這種圖片,加上 我愛中國 四個字 尺寸[4:3]" },
        { "type": "image_url", "image_url": { "url": "https://example.com/image.png" } }
      ]
    }
  ]
}

函數調用 / Function Calling

POST/v1/chat/completions

讓模型根據提示自動選擇並輸出函數調用 JSON。 / Allow the model to call functions.

Example Request

{
  "model": "gpt-4o",
  "messages": [{ "role": "user", "content": "北京今天天氣怎麼樣 / How is the weather in Beijing?" }],
  "tools": [{
    "type": "function",
    "function": {
      "name": "get_current_weather",
      "description": "Get the current weather in a given location",
      "parameters": {
        "type": "object",
        "properties": {
          "location": { "type": "string" },
          "unit": { "type": "string", "enum": ["celsius", "fahrenheit"] }
        },
        "required": ["location"]
      }
    }
  }]
}

結構化輸出 / Structured Outputs

POST/v1/chat/completions

強制模型按照指定的 JSON Schema 輸出格式。 / Force the model to output matching a JSON Schema.

Example Request

{
  "model": "gpt-4.5-preview",
  "messages": [{ "role": "user", "content": "How do I prepare for a job interview?" }],
  "response_format": {
    "type": "json_schema",
    "json_schema": {
      "name": "content_compliance",
      "schema": {
        "type": "object",
        "properties": {
          "is_violating": { "type": "boolean" },
          "category": { "type": "string" }
        },
        "required": ["is_violating", "category"],
        "additionalProperties": false
      },
      "strict": true
    }
  }
}

控制推理模型努力程度 / Reasoning Effort

POST/v1/chat/completions

調整思考型模型(如 o1, o3-mini)的推理深度。 / Control reasoning depth for thinking models.

Example Request

{
  "model": "o3-mini",
  "messages": [{ "role": "user", "content": "你好 / Hello" }],
  "reasoning_effort": "medium"
}

DeepSeek v3/R1 思考程度 / DeepSeek Reasoning

POST/v1/chat/completions

針對 DeepSeek 模型控制深度思考能力。 / Control deep thinking for DeepSeek models.

Example Request

{
  "model": "deepseek-reasoner",
  "messages": [{ "role": "user", "content": "你好 / Hello" }],
  "thinking": {
    "type": "enabled"
  }
}

DeepSeek OCR 識別 / DeepSeek OCR

POST/v1/chat/completions

利用 DeepSeek 進行精準的光學字元識別。 / Optical Character Recognition using DeepSeek.

Example Request

{
  "model": "deepseek-ocr",
  "stream": true,
  "messages": [
    { "role": "system", "content": "<image>\nFree OCR." },
    {
      "role": "user",
      "content": [
        { "type": "image_url", "image_url": { "url": "https://example.com/document.jpg" } }
      ]
    }
  ]
}

創建完成 / Completions (Legacy)

POST/v1/completions

傳統的文本補全端點。 / Legacy text completion endpoint.

Example Request

{
  "model": "gpt-3.5-turbo-instruct",
  "prompt": "你好, / Hello,",
  "max_tokens": 30,
  "temperature": 0
}

創建嵌入 / Create Embeddings

POST/v1/embeddings

獲取輸入文本的向量表示。 / Get vector representation of input text.

Example Request

{
  "model": "text-embedding-3-large",
  "input": "喵喵喵喵喵喵喵喵喵喵喵喵喵喵喵 / Meow meow meow"
}

列出模型 / List Models

GET/v1/models

獲取所有可用模型的列表。 / Get a list of all available models.

Example Request

// No Request Body Required

創建思考聊天 / Claude Extended Thinking

POST/v1/messages

Claude 模型的擴展思考模式調用。 / Claude extended thinking mode.

Example Request

{
  "model": "claude-sonnet-4-20250514",
  "system": "你是一个智能AI助手,叫小王",
  "messages": [
    { "role": "user", "content": "你是谁?!" }
  ],
  "stream": true,
  "max_tokens": 8000,
  "thinking": {
    "type": "enabled",
    "budget_tokens": 1200
  }
}

Claude 聊天補全 (流式) / Claude Chat (Stream)

POST/v1/chat/completions

Claude 模型的流式聊天補全。 / Claude streaming chat completions.

Example Request

{
  "model": "claude-3-7-sonnet-20250219",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "Hello!" }
  ],
  "stream": true
}

Claude 聊天補全 (非流) / Claude Chat (Non-stream)

POST/v1/chat/completions

Claude 模型的非流式聊天補全。 / Claude non-streaming chat completions.

Example Request

{
  "model": "claude-sonnet-4-20250514",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": "Hello!" }
  ]
}

Claude 聊天識圖 (流式) / Claude Vision (Stream)

POST/v1/chat/completions

Claude 模型的流式圖片理解。 / Claude vision streaming.

Example Request

{
  "model": "claude-sonnet-4-20250514",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": [
        { "type": "text", "text": "这张图片里有什么?请详细描述。" },
        { "type": "image_url", "image_url": { "url": "https://example.com/image.png" } }
    ]}
  ],
  "stream": true
}

Claude 聊天識圖 (非流) / Claude Vision (Non-stream)

POST/v1/chat/completions

Claude 模型的非流式圖片理解。 / Claude vision non-streaming.

Example Request

{
  "model": "claude-sonnet-4-20250514",
  "messages": [
    { "role": "system", "content": "You are a helpful assistant." },
    { "role": "user", "content": [
        { "type": "text", "text": "这张图片里有什么?请详细描述。" },
        { "type": "image_url", "image_url": { "url": "https://example.com/image.png" } }
    ]}
  ]
}

Gemini 文本生成 / Text Generation

POST/v1beta/models/gemini-2.5-pro:generateContent

Gemini 模型原生格式文本生成。 / Gemini native format text generation.

Example Request

{
  "systemInstruction": {
    "parts": [
      { "text": "你是一直小猪.你会在回复开始的时候 加一个'哼哼'" }
    ]
  },
  "contents": [
    {
      "role": "user",
      "parts": [
        { "text": "你是谁?" }
      ]
    }
  ],
  "generationConfig": {
    "temperature": 1,
    "topP": 1,
    "thinkingConfig": {
      "includeThoughts": true,
      "thinkingBudget": 26240
    }
  }
}

Gemini 文本生成 (流式) / Text Generation (Stream)

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

Gemini 模型原生格式流式輸出。 / Gemini native format streaming.

Example Request

{
  "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
    }
  }
}

Gemini 文本生成+思考 (流式) / Text+Thinking (Stream)

POST/v1beta/models/gemini-2.5-pro:streamGenerateContent

Gemini 結合深度思考的流式輸出。 / Gemini streaming with deep thinking.

Example Request

{
  "contents": [
    {
      "parts": [{ "text": "1+2+3+4+5+....+999?" }],
      "role": "user"
    }
  ],
  "systemInstruction": {
    "parts": [{ "text": "hi" }],
    "role": "user"
  },
  "generationConfig": {
    "temperature": 1,
    "topP": 1,
    "thinkingConfig": {
      "includeThoughts": true,
      "thinkingBudget": 26240
    }
  }
}

Gemini 圖片生成 / Image Generation

POST/v1beta/models/{modeName}:generateContent

Gemini 原生格式的文生圖與圖生圖。 / Gemini native text-to-image and image-to-image.

Example Request

{
  "contents": [
    {
      "role": "user",
      "parts": [
        { "text": "Hi, can you create a 3d rendered image of a pig with wings and a top hat flying over a happy futuristic scifi city with lots of greenery?" }
      ]
    }
  ],
  "generationConfig": {
    "responseModalities": ["IMAGE"]
  }
}