### cURL Example: Get File List

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/get-file-list

Example cURL command to retrieve a list of files from the SiliconFlow API, demonstrating the GET request method and required authorization header.

```cURL
curl --request GET \
  --url https://api.siliconflow.cn/v1/files \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### API Reference: Get File List Response Example

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/get-file-list

Illustrates the successful JSON response structure for the GET /files API endpoint, showing an example of a file object with its properties like ID, filename, purpose, and size.

```APIDOC
Response (200 application/json):
  description: Successful response with file list
  type: object
  example:
    code: 20000
    message: "Ok"
    status: true
    data:
      data:
        - id: "file-kkhtqklcnm"
          object: "file"
          bytes: 806
          created_at: 1741777570
          filename: "requests-2.jsonl"
          purpose: "batch"
          line_count: 2
      object: "file"
```

--------------------------------

### cURL Example for Retrieving Models List

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/models/get-model-list

A cURL command example demonstrating how to make a GET request to the /v1/models endpoint, including the necessary Authorization header with a Bearer token for authentication.

```cURL
curl --request GET \
  --url https://api.siliconflow.cn/v1/models \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### cURL Example: Create Video Generation Request

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/videos_submit

Demonstrates how to make a POST request to the /v1/video/submit endpoint using cURL, including necessary headers for authorization and content type, and a JSON body with example parameters for video generation.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/video/submit \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "Wan-AI/Wan2.1-T2V-14B",
  "prompt": "<string>",
  "negative_prompt": "<string>",
  "image_size": "1280x720",
  "image": "https://inews.gtimg.com/om_bt/Os3eJ8u3SgB3Kd-zrRRhgfR5hUvdwcVPKUTNO6O7sZfUwAA/641",
  "seed": 123
}'
```

--------------------------------

### Successful JSON Response Example for Models List

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/models/get-model-list

An example JSON response returned upon a successful GET request to the /v1/models endpoint, showing the structure of the model list object and a sample model entry.

```JSON
{
  "object": "list",
  "data": [
    {
      "id": "stabilityai/stable-diffusion-xl-base-1.0",
      "object": "model",
      "created": 0,
      "owned_by": ""
    }
  ]
}
```

--------------------------------

### Successful Video Generation Link API Response

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/get_videos_status

An example of the JSON response returned upon a successful request to retrieve a video generation link, detailing the structure of the `results` object including video URLs, timings, and seed.

```JSON
{
  "status": "Succeed",
  "reason": "<string>",
  "results": {
    "videos": [
      {
        "url": "<string>"
      }
    ],
    "timings": {
      "inference": 123
    },
    "seed": 123
  }
}
```

--------------------------------

### cURL Example: Retrieve Batch Details

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/get-batch

Example cURL command to send a GET request to the SiliconFlow API to retrieve the details of a specific batch task, requiring an authorization token.

```cURL
curl --request GET \
  --url https://api.siliconflow.cn/v1/batches/{batch_id} \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### Example cURL Request to Get Voice Styles List

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/voice-list

Demonstrates how to make a GET request to the SiliconFlow API's /v1/audio/voice/list endpoint using cURL. This request requires an 'Authorization' header with a Bearer token.

```cURL
curl --request GET \
  --url https://api.siliconflow.cn/v1/audio/voice/list \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### Qwen2.5-7B-Instruct System Prompt for 'Smart Sayings' Application (Before Fine-tuning)

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/userguide/guides/fine-tune

This prompt defines the role, task, input, and output format for the Qwen2.5-7B-Instruct model to generate 'golden phrases' in a critical and humorous style. It specifies the persona, provides examples, and enforces a strict JSON output format, resulting in a lengthy prompt.

```AI Prompt
# 角色
你是一位新潮评论家，你年轻、批判，又深刻；
你言辞犀利而幽默，擅长一针见血得表达隐喻，对现实的批判讽刺又不失文雅；
你的行文风格和"Oscar Wilde" "鲁迅" "林语堂"等大师高度一致；
从情感上要是对输入的否定。
# 任务
## 金句诠释
用特殊视角来全新得诠释给定的汉语词汇；
敏锐得抓住给定的词汇的本质，用“辛辣的讽刺”“一针见血的评论”的风格构造包含隐喻又直达本质的「金句」
例如：
"合伙人"： "一同下海捞金时，个个都是乘风破浪的水手，待到分金之际，方知彼此是劫财的海盗。"
"大数据"： "看似无所不能的数字神明，实则不过是现代社会的数字鸦片，让人沉溺于虚幻的精准，却忽略了人性的复杂与多变。"
"股市"： "万人涌入的淘金场，表面上是财富的摇篮，实则多数人成了填坑的沙土。"
"白领"： "西装革履，看似掌握命运的舵手，实则不过是写字楼里的高级囚徒。"
"金融家"： "在金钱的海洋中遨游，表面上是操纵风浪的舵手，实则不过是随波逐流的浮萍。"
"城市化"： "乡村的宁静被钢铁森林吞噬，人们在追逐繁华的幻影中，遗失了心灵的田园。"
"逃离北上广"： "逃离繁华的都市牢笼，看似追逐自由的灵魂，实则不过是换个地方继续画地为牢。"
"基金"： "看似为财富增值保驾护航的金融巨轮，实则多数人不过是随波逐流的浮萍，最终沦为填补市场波动的牺牲品。"
# 输入
用户直接输入词汇。
# 输出
严格输出JSON格式，包括两个字段，“prompt”为用户的输入；“output”为用户的金句内容，不额外输出额外任何其他内容，不要输出引号，严格限制用户的输入的词汇绝对不能出现在输出中，注意突出转折和矛盾，输出内容为一句话，最后以“。”结束，中间的停顿使用“，”分隔。例如
{
  "prompt": "合伙人",
  "output": "一同下海捞金时，个个都是乘风破浪的水手，待到分金之际，方知彼此是劫财的海盗。"
}
```

--------------------------------

### JSON Example: Successful Video Generation Response

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/videos_submit

Illustrates the expected JSON response structure upon a successful video generation request, containing the unique request ID.

```JSON
{
  "requestId": "<string>"
}
```

--------------------------------

### API Reference: Get User-Defined Voice Styles List

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/voice-list

Detailed API documentation for the GET /audio/voice/list endpoint, including method, path, authorization requirements, and the structure of a successful JSON response.

```APIDOC
Endpoint:
  Method: GET
  Path: /audio/voice/list
  Description: Get list of user-defined voice styles

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer <your api key>

Responses:
  200 OK:
    Content-Type: application/json
    Description: Successful response, User-defined voice style list response
    Body:
      {
        "results": [
          {
            "model": "fishaudio/fish-speech-1.4",
            "customName": "your-voice-name",
            "text": "在一无所知中, 梦里的一天结束了，一个新的轮回便会开始",
            "uri": "speech:your-voice-name:xxx:xxx"
          }
        ]
      }
```

--------------------------------

### Example Chat Completion API Response

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/chat-completions/chat-completions

Illustrates a successful JSON response from the SiliconFlow chat completions API, detailing the structure of choices, message content, tool calls, usage statistics, and metadata like ID, creation timestamp, model, and object type.

```JSON
{
  "id": "<string>",
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "<string>",
        "reasoning_content": "<string>",
        "tool_calls": [
          {
            "id": "<string>",
            "type": "function",
            "function": {
              "name": "<string>",
              "arguments": "<string>"
            }
          }
        ]
      },
      "finish_reason": "stop"
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  },
  "created": 123,
  "model": "<string>",
  "object": "chat.completion"
}
```

--------------------------------

### Get Video Generation Link Request with cURL

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/get_videos_status

Example cURL command to send a POST request to the /v1/video/status endpoint, including authorization and the requestId in the JSON body.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/video/status \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "requestId": "<string>"
}'
```

--------------------------------

### cURL Request to Create Audio Transcription

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/create-audio-transcriptions

Example cURL command demonstrating how to send a POST request to the audio transcription endpoint, including required headers for authorization and content type, and specifying the model.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/audio/transcriptions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form model=FunAudioLLM/SenseVoiceSmall
```

--------------------------------

### cURL Example: Create Image Generation Request

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

This cURL command demonstrates how to send a POST request to the SiliconFlow image generation API. It includes necessary headers for authorization and content type, along with a JSON payload specifying the model, prompt, image size, and other generation parameters.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/images/generations \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "Kwai-Kolors/Kolors",
  "prompt": "an island near sea, with seagulls, moon shining over the sea, light house, boats int he background, fish flying over the sea",
  "negative_prompt": "<string>",
  "image_size": "1024x1024",
  "batch_size": 1,
  "seed": 4999999999,
  "num_inference_steps": 20,
  "guidance_scale": 7.5,
  "image": "data:image/png;base64, XXX"
}'
```

--------------------------------

### Upload Reference Audio Success Response

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/upload-voice

Example JSON response returned upon successful upload of a reference audio file, containing the URI for the newly uploaded voice style.

```json
{
  "uri": "speech:your-voice-name:xxx:xxx"
}
```

--------------------------------

### Upload Reference Audio using cURL

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/upload-voice

Example cURL command to upload a reference audio file to the SiliconFlow API, including authorization, content type, and form data for audio, model, custom name, and associated text.

```curl
curl --request POST \
  --url https://api.siliconflow.cn/v1/uploads/audio/voice \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form 'audio=data:audio/mpeg;base64,aGVsbG93b3JsZA==' \
  --form model=FunAudioLLM/CosyVoice2-0.5B \
  --form customName=your-voice-name \
  --form 'text=在一无所知中, 梦里的一天结束了，一个新的轮回便会开始'
```

--------------------------------

### API Reference: POST /video/status

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/get_videos_status

Comprehensive API documentation for the POST /video/status endpoint, detailing authentication methods, required request body parameters, and the structure of the successful 200 JSON response.

```APIDOC
Method: POST
Path: /video/status

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Description: Use the following format for authentication: Bearer [<your api key>]

Request Body (application/json):
  Properties:
    requestId:
      Type: string
      Description: The ID of the video generation request.

Response (200 - application/json):
  Description: Successful response
  Type: object
  Properties:
    status: string
    reason: string
    results:
      Type: object
      Properties:
        videos:
          Type: array
          Items:
            Type: object
            Properties:
              url: string
        timings:
          Type: object
          Properties:
            inference: number
        seed: number
```

--------------------------------

### API Reference: Get File List Request Details

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/get-file-list

Documents the GET /files API endpoint, outlining the request method, path, required authorization header, and available query parameters for filtering and limiting results.

```APIDOC
GET /files

Authorizations:
  Authorization:
    type: string
    in: header
    required: true
    description: Use the following format for authentication: Bearer [<your api key>]

Query Parameters:
  purpose:
    type: enum<string>
    required: true
    description: Filter files by their purpose
    available_options: ["batch"]
  limit:
    type: integer
    default: 10
    description: Maximum number of files to return (default 10)
    required_range: "x >= 1"
```

--------------------------------

### Example Embedding API Response

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/embeddings/create-embeddings

Illustrates the expected JSON response structure for a successful embedding request, including the model used, an array of embedding objects, and usage statistics.

```JSON
{
  "model": "<string>",
  "data": [
    {
      "object": "embedding",
      "embedding": [
        123
      ],
      "index": 123
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  }
}
```

--------------------------------

### API Reference: Get Batch Details Endpoint

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/get-batch

Detailed API documentation for the GET /batches/{batch_id} endpoint, outlining authorization, path parameters, and the structure of the successful JSON response for a batch object.

```APIDOC
GET /batches/{batch_id}

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer <your api key>

Path Parameters:
  batch_id:
    Type: string
    Required: true
    Description: The ID of the batch to retrieve.

Response (200 OK):
  Content-Type: application/json
  Description: The Batch object matching the specified ID.
  Schema: object
  Example:
    {
      "id": "batch_rdyqgrcgjg",
      "object": "batch",
      "endpoint": "/v1/chat/completions",
      "errors": [],
      "input_file_id": "file-jkvytbjtow",
      "completion_window": "24h",
      "status": "in_queue",
      "output_file_id": null,
      "error_file_id": null,
      "created_at": 1741685413,
      "in_progress_at": null,
      "expires_at": 1741771813,
      "finalizing_at": null,
      "completed_at": null,
      "failed_at": null,
      "expired_at": null,
      "cancelling_at": null,
      "cancelled_at": null,
      "request_counts": {},
      "metadata": {
        "description": "nightly eval job"
      }
    }
```

--------------------------------

### Get User Account Information API Endpoint

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/userinfo/get-user-info

Documents the API endpoint for retrieving user account details, including balance and status. It specifies the GET method, the /user/info path, required authorization, and the structure of a successful JSON response.

```APIDOC
Endpoint: GET /user/info
Description: Get user information including balance and status

Authorizations:
  Authorization: string (header, required)
    Description: Use the following format for authentication: Bearer <your api key>
    Source: https://cloud.siliconflow.cn/account/ak

Response (200 OK):
  Content-Type: application/json
  Description: Successful response
  Schema:
    {
      "code": 20000,
      "message": "OK",
      "status": true,
      "data": {
        "id": "userid",
        "name": "username",
        "image": "user_avatar_image_url",
        "email": "user_email_address",
        "isAdmin": false,
        "balance": "0.88",
        "status": "normal",
        "introduction": "",
        "role": "",
        "chargeBalance": "88.00",
        "totalBalance": "88.88"
      }
    }
```

```cURL
curl --request GET \
  --url https://api.siliconflow.cn/v1/user/info \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### cURL Example: Cancel SiliconFlow Batch Task

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/cancel-batch

Example cURL command to send a POST request to cancel a batch task using its ID and an authorization token. This command demonstrates how to interact with the SiliconFlow API for batch management.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/batches/{batch_id}/cancel \
  --header 'Authorization: Bearer <token>'
```

--------------------------------

### JSON Response Example: Successful Batch Cancellation

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/cancel-batch

An example of the JSON object returned upon successful cancellation of a batch task, detailing its ID, object type, associated endpoint, input file, completion window, current status, and metadata.

```JSON
{
  "id": "batch_rdyqgrcgjg",
  "object": "batch",
  "endpoint": "/v1/chat/completions",
  "errors": null,
  "input_file_id": "file-jkvytbjtow",
  "completion_window": "24h",
  "status": "in_queue",
  "output_file_id": null,
  "error_file_id": null,
  "created_at": 1741685413,
  "in_progress_at": null,
  "expires_at": 1741771813,
  "finalizing_at": null,
  "completed_at": null,
  "failed_at": null,
  "expired_at": null,
  "cancelling_at": null,
  "cancelled_at": null,
  "request_counts": null,
  "metadata": {
    "description": "nightly eval job"
  }
}
```

--------------------------------

### Create Rerank Request using cURL

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/rerank/create-rerank

Example cURL command to send a POST request to the /v1/rerank endpoint. It demonstrates how to include the Authorization header and the JSON payload for reranking documents based on a query.

```curl
curl --request POST \
  --url https://api.siliconflow.cn/v1/rerank \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "BAAI/bge-reranker-v2-m3",
  "query": "Apple",
  "documents": [
    "apple",
    "banana",
    "fruit",
    "vegetable"
  ],
  "top_n": 4,
  "return_documents": false,
  "max_chunks_per_doc": 1024,
  "overlap_tokens": 80
}'
```

--------------------------------

### Retrieve Models List API Endpoint Documentation

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/models/get-model-list

Detailed documentation for the GET /models API endpoint, outlining its purpose, required authentication, available query parameters for filtering models, and the expected structure of a successful response.

```APIDOC
Endpoint: GET /models
Description: Retrieve models information.

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer <your api key>

Query Parameters:
  type:
    Type: enum<string>
    Description: The type of models
    Available options: text, image, audio, video
  sub_type:
    Type: enum<string>
    Description: The sub type of models. You can use it to filter models individually without setting type.
    Available options: chat, embedding, reranker, text-to-image, image-to-image, speech-to-text, text-to-video

Response:
  Status: 200 OK
  Content-Type: application/json
  Description: Successful response
  Schema: object
```

--------------------------------

### Qwen2.5-7B-Instruct System Prompt for 'Smart Sayings' Application (After Fine-tuning)

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/userguide/guides/fine-tune

This simplified prompt is used with the Qwen2.5-7B-Instruct model after fine-tuning with the 'Smart Sayings' corpus. The fine-tuning significantly reduces the prompt length from 553 tokens to 8 tokens, improving speed and optimizing cost while maintaining or enhancing output quality and consistency.

```AI Prompt
你是智说新语生成器
```

--------------------------------

### SiliconFlow Chat Completions API Reference

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/chat-completions/chat-completions

Comprehensive API documentation for the SiliconFlow chat completions endpoint, detailing required authorization headers and the structure of the request body, including supported parameters for LLM and VLM models.

```APIDOC
Endpoint: POST /chat/completions

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Description: Use the following format for authentication: Bearer [<your api key>]
    Link: https://cloud.siliconflow.cn/account/ak

Body:
  Content-Type: application/json
  Supported Models: LLM, VLM
```

--------------------------------

### Google Search Tool Call by AI Agent

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/usercases/use-siliconcloud-in-openmanus

A JSON-formatted tool call demonstrating how an AI agent invokes the `google_search` function with specific arguments to find relevant reports on the global soybean market for 2025.

```JSON
<tool_call>
  "name": "google_search",
  "arguments": {
    "query": "global soybean market analysis 2025 report",
    "num_results": 10
  }
</tool_call>
```

--------------------------------

### API Reference: Create Video Generation Request

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/videos/videos_submit

Details the POST /video/submit API endpoint for generating videos, including request parameters, authorization requirements, and expected response format. Users need to poll a separate status interface to retrieve the video link, which is valid for 10 minutes.

```APIDOC
Endpoint: POST /video/submit
Description: Generate a video through the input prompt. This API returns the user’s current request ID. The user needs to poll the status interface to get the specific video link. The generated result is valid for 10 minutes, so please retrieve the video link promptly.

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer [<your api key>]

Request Body (application/json):
  model: string (e.g., "Wan-AI/Wan2.1-T2V-14B")
  prompt: string
  negative_prompt: string
  image_size: string (e.g., "1280x720")
  image: string (URL, e.g., "https://inews.gtimg.com/om_bt/Os3eJ8u3SgB3Kd-zrRRhgfR5hUvdwcVPKUTNO6O7sZfUwAA/641")
  seed: integer (e.g., 123)

Response (200 OK - application/json):
  requestId: string
```

--------------------------------

### SiliconFlow Rerank API Reference

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/rerank/create-rerank

Comprehensive API documentation for the POST /v1/rerank endpoint, detailing request parameters, authorization, and the expected response structure for creating a rerank request.

```APIDOC
POST /v1/rerank

Headers:
  Authorization: Bearer <token> (string, header, required)
    Description: Use the following format for authentication: Bearer [<your api key>]

Request Body (application/json):
  model: string (required)
    Example: "BAAI/bge-reranker-v2-m3"
  query: string (required)
    Description: The query string.
    Example: "Apple"
  documents: array of string (required)
    Description: List of documents to rerank.
    Example: ["apple", "banana", "fruit", "vegetable"]
  top_n: integer (optional)
    Description: Number of top results to return.
    Example: 4
  return_documents: boolean (optional)
    Description: Whether to return the full documents in the response.
    Example: false
  max_chunks_per_doc: integer (optional)
    Description: Maximum chunks per document.
    Example: 1024
  overlap_tokens: integer (optional)
    Description: Overlap tokens.
    Example: 80

Response (200 OK, application/json):
  id: string
    Description: Unique identifier for the request.
    Example: "<string>"
  results: array of object
    document: object
      text: string
        Description: The text of the reranked document.
        Example: "<string>"
    index: integer
      Description: Original index of the document.
      Example: 123
    relevance_score: number
      Description: The relevance score of the document.
      Example: 123
  tokens: object
    input_tokens: integer
      Description: Number of input tokens.
      Example: 123
    output_tokens: integer
      Description: Number of output tokens.
      Example: 123

Possible HTTP Status Codes:
  200: OK
  400: Bad Request
  401: Unauthorized
  404: Not Found
  429: Too Many Requests
  503: Service Unavailable
  504: Gateway Timeout
```

--------------------------------

### Image Generation API Endpoint Overview

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

Defines the HTTP method and path for the SiliconFlow image generation API. This endpoint allows users to create images based on provided text prompts.

```APIDOC
POST /images/generations
```

--------------------------------

### Integrate translate.js with SiliconFlow Service

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/usercases/use-siliconcloud-in-translate-js

This JavaScript snippet demonstrates how to quickly integrate the translate.js library into an HTML page and configure it to use SiliconFlow's translation service. It should be placed at the bottom of the HTML body for optimal loading.

```HTML
<script src="https://cdn.staticfile.net/translate.js/3.15.6/translate.min.js"></script>
<script>
translate.service.use('siliconflow'); //设置采用硅基流动的翻译通道
translate.execute(); //触发翻译
</script>
```

--------------------------------

### API Authorization Details

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

Provides details on the required Authorization header for authenticating requests to the SiliconFlow API. It specifies the format for including the API key.

```APIDOC
Authorization:
  Type: string
  Location: header
  Required: true
  Format: Bearer [<your api key>]
  Link: https://cloud.siliconflow.cn/account/ak
```

--------------------------------

### Embeddings API Endpoint Reference

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/embeddings/create-embeddings

Detailed API documentation for the /v1/embeddings endpoint, covering the POST method, authorization requirements, request body structure, and possible response statuses.

```APIDOC
POST /embeddings
Description: Creates an embedding vector representing the input text.

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer [<your api key>] (https://cloud.siliconflow.cn/account/ak)

Body:
  Content-Type: application/json
  Fields:
    model: string (e.g., "BAAI/bge-large-zh-v1.5")
    input: string (e.g., "Silicon flow embedding online: fast, affordable, and high-quality embedding services. come try it out!")
    encoding_format: string (e.g., "float")

Response:
  200 OK:
    Description: The response is of type `object`.
    Content-Type: application/json
    Schema:
      model: string
      data: array of objects
        object: string (e.g., "embedding")
        embedding: array of numbers (e.g., [123])
        index: number (e.g., 123)
      usage: object
        prompt_tokens: number (e.g., 123)
        completion_tokens: number (e.g., 123)
        total_tokens: number (e.g., 123)
  Error Codes:
    400: Bad Request
    401: Unauthorized
    404: Not Found
    429: Too Many Requests
    503: Service Unavailable
    504: Gateway Timeout
```

--------------------------------

### API Endpoint: Create Audio Transcription

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/create-audio-transcriptions

Defines the HTTP method and path for creating an audio transcription request via the SiliconFlow API.

```APIDOC
POST /audio/transcriptions
```

--------------------------------

### Create Chat Completion Request with cURL

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/chat-completions/chat-completions

Demonstrates how to send a POST request to the SiliconFlow chat completions API using cURL, including authorization and a JSON request body with parameters like model, messages, stream, max_tokens, and tools.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "Qwen/QwQ-32B",
  "messages": [
    {
      "role": "user",
      "content": "What opportunities and challenges will the Chinese large model industry face in 2025?"
    }
  ],
  "stream": false,
  "max_tokens": 512,
  "enable_thinking": false,
  "thinking_budget": 4096,
  "min_p": 0.05,
  "stop": null,
  "temperature": 0.7,
  "top_p": 0.7,
  "top_k": 50,
  "frequency_penalty": 0.5,
  "n": 1,
  "response_format": {
    "type": "text"
  },
  "tools": [
    {
      "type": "function",
      "function": {
        "description": "<string>",
        "name": "<string>",
        "parameters": {},
        "strict": false
      }
    }
  ]
}'
```

--------------------------------

### API Authorization Header Specification

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/create-audio-transcriptions

Details the required 'Authorization' header for API requests, specifying its type, location, and the 'Bearer' token format.

```APIDOC
Authorization
string
header
required
Use the following format for authentication: Bearer [<your api key>](https://cloud.siliconflow.cn/account/ak)
```

--------------------------------

### SiliconFlow Upload Reference Audio API Specification

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/upload-voice

Detailed specification for the POST /uploads/audio/voice endpoint, including authorization, request body parameters, and successful response structure.

```APIDOC
Endpoint: POST /uploads/audio/voice

Description: Upload user-provided voice style, which can be in base64 encoding or file format.

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Format: Bearer <your api key>

Request Body: multipart/form-data
  Parameters (implied from cURL example):
    audio: string (base64 encoded audio data, e.g., data:audio/mpeg;base64,aGVsbG93b3JsZA==)
    model: string (e.g., FunAudioLLM/CosyVoice2-0.5B)
    customName: string (e.g., your-voice-name)
    text: string (e.g., 在一无所知中, 梦里的一天结束了，一个新的轮回便会开始)

Responses:
  200 OK:
    Content-Type: application/json
    Description: Successful response
    Body:
      uri: string (e.g., speech:your-voice-name:xxx:xxx)

  Error Codes: 400, 401, 404, 429, 503, 504
```

--------------------------------

### Audio Transcription API Success Response (200 OK)

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/create-audio-transcriptions

Describes the successful 200 OK response for audio transcription, including its content type and the JSON schema for the returned data.

```APIDOC
200
application/json
Represents a transcription response returned by model, based on the provided input.
{
  "text": "<string>"
}
```

--------------------------------

### Upload Files API Reference

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/upload-file

Documents the API endpoint for uploading files to SiliconFlow, primarily for batch processing. This endpoint supports `multipart/form-data` and requires `Bearer` token authentication. It details the request method, URL, authorization, body parameters, and possible response statuses.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/files \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: multipart/form-data' \
  --form purpose=batch
```

```APIDOC
Endpoint: POST /v1/files
Purpose: Upload files for batch processing.

Authorizations:
  Authorization: string (header, required)
    Format: Bearer <your api key>

Body:
  Content-Type: multipart/form-data
  Parameters:
    purpose: string (form data, required)
      Value: batch

Response Status Codes:
  200: OK
  400: Bad Request
  401: Unauthorized
  404: Not Found
  503: Service Unavailable
  504: Gateway Timeout
```

```JSON
{
  "code": 20000,
  "message": "Ok",
  "status": true,
  "data": {
    "id": "file-jkvytbjtow",
    "object": "file",
    "bytes": 8509,
    "createdAt": 1741685396,
    "filename": "requests.jsonl",
    "purpose": "batch"
  }
}
```

--------------------------------

### API Request Body Format

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/create-audio-transcriptions

Specifies that the request body for audio transcription must be sent as 'multipart/form-data'.

```APIDOC
multipart/form-data
```

--------------------------------

### API Response Schema: Successful Image Generation (200 OK)

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

Describes the JSON structure returned upon a successful image generation request (HTTP 200 OK). It includes an array of generated image URLs, inference timing details, and the seed used for the generation process.

```APIDOC
{
  "images": [
    {
      "url": "<string>"
    }
  ],
  "timings": {
    "inference": 123
  },
  "seed": 123
}
```

--------------------------------

### Create Embedding Request with cURL

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/embeddings/create-embeddings

Demonstrates how to send a POST request to the /v1/embeddings endpoint using cURL, including authorization and JSON payload for model, input text, and encoding format.

```cURL
curl --request POST \
  --url https://api.siliconflow.cn/v1/embeddings \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "model": "BAAI/bge-large-zh-v1.5",
  "input": "Silicon flow embedding online: fast, affordable, and high-quality embedding services. come try it out!",
  "encoding_format": "float"
}'
```

--------------------------------

### API Response Status Codes

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

Lists the various HTTP status codes that can be returned by the image generation API, indicating the outcome of the request, from success to different error conditions.

```APIDOC
200: OK
400: Bad Request
401: Unauthorized
404: Not Found
429: Too Many Requests
503: Service Unavailable
504: Gateway Timeout
```

--------------------------------

### API Request Body Content Type

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/images/images-generations

Specifies the required Content-Type header for the request body when submitting data to the image generation API.

```APIDOC
application/json
```

--------------------------------

### API Reference: Cancel Batch Task Endpoint

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/batch/cancel-batch

Detailed API documentation for canceling a batch task, including the endpoint, required authorization, path parameters, and possible HTTP response status codes.

```APIDOC
Endpoint: POST /batches/{batch_id}/cancel

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Description: Use the format: Bearer [<your api key>]

Path Parameters:
  batchId:
    Type: string
    Required: true
    Description: Unique identifier of the batch to cancel

Response Status Codes:
  200: Successful response
  400: Bad Request
  401: Unauthorized
  404: Not Found
  503: Service Unavailable
  504: Gateway Timeout
```

--------------------------------

### Delete Reference Audio API Request

Source: https://docs.siliconflow.cn/cn/api-reference/chat-completions/chat-completions/api-reference/audio/delete-voice

This API endpoint allows users to delete a previously uploaded user-defined voice style by providing its unique URI. Authentication is required using a Bearer token.

```curl
curl --request POST \
  --url https://api.siliconflow.cn/v1/audio/voice/deletions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{
  "uri": "speech:your-voice-name:xxx:xxxx"
}'
```

```APIDOC
Endpoint: POST /audio/voice/deletions

Description: Delete user-defined voice style

Authorizations:
  Authorization:
    Type: string
    Location: header
    Required: true
    Description: Use the following format for authentication: Bearer [<your api key>]

Request Body (application/json):
  Parameters:
    uri:
      Type: string
      Description: The unique URI of the voice style to delete.
      Example: "speech:your-voice-name:xxx:xxxx"

Response (200 - Successful response):
  Content-Type: application/json
  Body: "<string>" (Indicates successful deletion, typically an empty string or confirmation message)

Possible HTTP Status Codes:
  200: OK - Successful response
  400: Bad Request
  401: Unauthorized
  404: Not Found
  429: Too Many Requests
  503: Service Unavailable
  504: Gateway Timeout
```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.