### GET /

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/segmenter.html

Counts tokens for the provided content via a simple GET request.

```APIDOC
## GET /

### Description
Counts the number of tokens in the provided content string.

### Method
GET

### Endpoint
https://segment.jina.ai/

### Parameters
#### Query Parameters
- **content** (string) - Required - The text content to count tokens for.

### Response
#### Success Response (200)
- **num_tokens** (integer) - The total count of tokens in the content.
- **num_chunks** (integer) - The number of chunks identified.
- **chunk_positions** (array) - The positions of the chunks.
- **chunks** (array) - The segmented text chunks.
```

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

### Send GET Request to Jina AI Search API

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Demonstrates how to send a GET request to the Jina AI search endpoint. Includes setting up the endpoint, headers, parameters, and data payload. Ensure you replace '<token here>' with your actual authentication token.

```python
endpoint = f"https://s.jina.ai/When%20was%20Jina%20AI%20founded?" # Base endpoint, required
headers = {
    "Accept": "application/json", # Response format, required
    "Authorization": "Bearer <token here>", # Authentication token, required
    "Content-Type": "application/json", # Request body format, optional
    "X-Custom-Header": "value" # Example of a custom header, optional
}
params = {
    "param_key": "param_value", # Example URL parameter, optional
}
data = {
    "request_key": "request_value", # Example request body data, optional
}
response = requests.get(endpoint, headers=headers, params=params, json=data) # Sends a GET request
```

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

### Perform a Search Request

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/s.reader.html

Examples of how to query the search endpoint using different programming languages.

```bash
curl 'https://s.jina.ai/When%20was%20Jina%20AI%20founded?' \
	-H "Accept: application/json" \
	-H "Authorization: Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq"
```

```javascript
fetch('https://s.jina.ai/When%20was%20Jina%20AI%20founded?', {
  method: 'GET',
  headers: {
    "Authorization": "Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq",
    "Accept": "application/json"
  },
})
```

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

### Segmenter API POST Response Example

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/segmenter.html

Example of a successful response from the Segmenter API when requesting tokens and chunks, including token details and chunk positions.

```json
{
      "num_tokens": 78,
      "tokenizer": "cl100k_base",
      "usage": {
        "tokens": 0
      },
      "num_chunks": 4,
      "chunk_positions": [
        [
          3,
          55
        ],
        [55,
          93
        ],
        [
          93,
          110
        ],
        [
          110,
          135
        ]
      ],
      "tokens": [
        [
          [
            "J",
            [
              41
            ]
          ],
          [
            "ina",
            [2259]
          ],
          [
            " AI",
            [15592]
          ],
          [
            ":",
            [25]
          ],
          [
            " Your",
            [4718]
          ],
          [
            " Search",
            [7694]
          ],
          [
            " Foundation",
            [5114]
          ],
          [
            ",",
            [11]
          ],
          [
            " Super",
            [7445]
          ],
          [
            "charged",
            [38061]
          ],
          [
            "!",
            [0]
          ],
          [
            " ",
            [11410]
          ],
          [
            "🚀",
            [248, 222]
          ],
          [
            "\n",
            [198]
          ],
          [
            "  ",
            [256]
          ]
        ],
        [

```

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

### GET /s.jina.ai/{query}

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/s.reader.html

Searches the web for a given query and returns URLs and content in an LLM-friendly format.

```APIDOC
## GET /s.jina.ai/{query}

### Description
Searches the web for a query and returns results in clean, LLM-friendly text.

### Method
GET

### Endpoint
https://s.jina.ai/{query}

### Parameters
#### Path Parameters
- **query** (string) - Required - The search query string.

### Request Example
curl 'https://s.jina.ai/When%20was%20Jina%20AI%20founded?' \
	-H "Accept: application/json" \
	-H "Authorization: Bearer <YOUR_API_KEY>"

### Response
#### Success Response (200)
- **code** (integer) - Status code
- **status** (integer) - Internal status code
- **data** (array) - List of search results

#### Response Example
{
  "code": 200,
  "status": 20000,
  "data": [
    {
      "title": "Jina AI - Your Search Foundation, Supercharged."
    }
  ]
}
```

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

### GET /https://{url}

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/r.reader.html

Fetches the content of a specified URL and returns it in a structured format suitable for LLM consumption.

```APIDOC
## GET /https://{url}

### Description
Fetches the content of a specified URL and returns it in a structured format.

### Method
GET

### Endpoint
https://r.jina.ai/{url}

### Parameters
#### Path Parameters
- **url** (string) - Required - The URL to be fetched and parsed.

### Request Example
curl 'https://r.jina.ai/https://example.com' \
	-H "Accept: application/json" \
	-H "Authorization: Bearer <YOUR_API_KEY>"

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code
- **status** (integer) - Internal status code
- **data** (object) - The parsed content object
  - **title** (string) - Page title
  - **url** (string) - The original URL
  - **content** (string) - The extracted text content
  - **usage** (object) - Token usage statistics

#### Response Example
{
  "code": 200,
  "status": 20000,
  "data": {
    "title": "Example Domain",
    "url": "https://example.com/",
    "content": "This domain is for use in illustrative examples...",
    "usage": {
      "tokens": 42
    }
  }
}
```

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

### Count Tokens with GET Request

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/segmenter.html

Use a GET request to the segment.jina.ai endpoint to count tokens in a given content string. The 'content' parameter should be URL-encoded.

```bash
https://segment.jina.ai/?content=
```

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

### Meta-Prompt API

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

This endpoint allows you to query information using natural language. It supports GET requests and requires authentication.

```APIDOC
## GET /jina-ai/meta-prompt

### Description
This endpoint is used to query information using natural language prompts. It returns structured JSON data or plain text depending on the request.

### Method
GET

### Endpoint
`https://s.jina.ai/<your_prompt>`

### Parameters
#### Query Parameters
- **param_key** (string) - Optional - Example URL parameter.

#### Headers
- **Accept**: `application/json` - Response format, required.
- **Authorization**: `Bearer <token here>` - Authentication token, required.
- **Content-Type**: `application/json` - Request body format, optional.
- **X-Custom-Header**: `value` - Example of a custom header, optional.

### Request Body
- **request_key** (string) - Optional - Example request body data.

### Request Example
```json
{
  "request_key": "request_value"
}
```

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code indicating success.
- **status** (integer) - Application-specific status code.
- **data** (array) - An array of objects containing the response details.
  - **title** (string) - The title of the response item.
  - **description** (string) - A description of the response item.
  - **url** (string) - A URL associated with the response item.
  - **content** (string) - Detailed content of the response.
  - **usage** (object) - Usage statistics.
    - **tokens** (integer) - Number of tokens used.

#### Response Example (JSON)
```json
{
  "code": 200,
  "status": 20000,
  "data": [
    {
      "title": "Jina AI - Your Search Foundation, Supercharged.",
      "description": "Description of the response",
      "url": "https://jina.ai/",
      "content": "Detailed content of the response ...",
      "usage": {
        "tokens": 11091
      }
    }
  ]
}
```

#### Response Example (Text)
Plain text response, potentially containing key-value pairs or summaries.
```

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

### GET /g.jina.ai/{query}

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/g.reader.html

Performs fact-checking on a provided query string.

```APIDOC
## GET /g.jina.ai/{query}

### Description
Performs fact-checking on a provided query string by verifying it against web references.

### Method
GET

### Endpoint
https://g.jina.ai/{query}

### Parameters
#### Path Parameters
- **query** (string) - Required - The statement or query to be fact-checked.

### Request Example
curl 'https://g.jina.ai/Jina%20AI%20was%20founded%20in%202020%20in%20Berlin.' \
	-H "Accept: application/json" \
	-H "Authorization: Bearer <your_token>"

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code
- **status** (integer) - Internal status code
- **data** (object) - Contains factuality, result, reason, references, and usage tokens.

#### Response Example
{
  "code": 200,
  "status": 20000,
  "data": {
    "factuality": 1,
    "result": true,
    "reason": "The statement is supported by multiple references.",
    "references": [
      {
        "url": "https://medium.com/jina-ai/2020-year-in-review-4896f7208fb0",
        "keyQuote": "Jina AI was founded in February 2020...",
        "isSupportive": true
      }
    ],
    "usage": {
      "tokens": 7520
    }
  }
}
```

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

### Grounding API JSON Response Example

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/g.reader.html

This is an example of a successful JSON response from the grounding API. It includes the factuality score, result, reasoning, supporting references, and token usage.

```json
{
      "code": 200,
      "status": 20000,
      "data": {
        "factuality": 1,
        "result": true,
        "reason": "The statement that Jina AI was founded in 2020 in Berlin is supported by multiple references. The first reference confirms the founding year as 2020, while the subsequent references explicitly state that Jina AI is based in Berlin, Germany. Therefore, both components of the statement are corroborated by the available evidence, making it factually correct.",
        "references": [
          {
            "url": "https://medium.com/jina-ai/2020-year-in-review-4896f7208fb0",
            "keyQuote": "Jina AI was founded in February 2020, in the midst of a global pandemic and economic slowdown.",
            "isSupportive": true
          },
          {
            "url": "https://www.cbinsights.com/company/jina-ai",
            "keyQuote": "It was founded in 2020 and is based in Berlin, Germany.",
            "isSupportive": true
          },
          {
            "url": "https://www.jinaai.cn/about-us/",
            "keyQuote": "Founded in 2020 and based in Berlin, Germany, Jina AI has swiftly risen as a leader in multimodal AI",
            "isSupportive": true
          },
          {
            "url": "https://www.linkedin.com/company/jinaai",
            "keyQuote": "Founded in February 2020, Jina AI has swiftly emerged as a global pioneer in multimodal AI technology.",
            "isSupportive": true
          }
        ],
        "usage": {
          "tokens": 7520
        }
      }
    }
```

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

### API Response Structure

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/reranker.html

Example of a successful 200 OK response from the Reranker API, containing the model name, usage statistics, and ranked results.

```json
{
  "model": "jina-reranker-v2-base-multilingual",
  "usage": {
    "total_tokens": 815
  },
  "results": [
    {
      "index": 0,
      "document": {
        "text": "Organic skincare for sensitive skin with aloe vera and chamomile: Imagine the soothing embrace of nature with our organic skincare range, crafted specifically for sensitive skin. Infused with the calming properties of aloe vera and chamomile, each product provides gentle nourishment and protection. Say goodbye to irritation and hello to a glowing, healthy complexion."
      },
      "relevance_score": 0.8783142566680908
    },
    {
      "index": 6,
      "document": {
        "text": "针对敏感肌专门设计的天然有机护肤产品：体验由芦荟和洋甘菊提取物带来的自然呵护。我们的护肤产品特别为敏感肌设计，温和滋润，保护您的肌肤不受刺激。让您的肌肤告别不适，迎来健康光彩。"
      }
```

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

### POST /v1/train

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/classifier-train-text.html

This endpoint allows you to train a custom classifier using provided text examples and labels. You can specify the model, access level, number of training iterations, and the input data.

```APIDOC
## POST /v1/train

### Description
Trains a custom classifier with provided text examples and labels.

### Method
POST

### Endpoint
https://api.jina.ai/v1/train

### Parameters
#### Request Body
- **model** (string) - Required - The name of the model to use for training (e.g., "jina-embeddings-v3").
- **access** (string) - Optional - Specifies the access level for the classifier. Can be "public" or "private". Defaults to "private".
- **num_iters** (integer) - Optional - Controls the training intensity. Higher values improve accuracy but increase token cost. Defaults to 10.
- **input** (array) - Required - An array of training examples, where each example is an object with "text" and "label" fields.
  - **text** (string) - Required - The input text for the training example.
  - **label** (string) - Required - The label associated with the input text.

### Request Example
```json
{
  "model": "jina-embeddings-v3",
  "access": "private",
  "num_iters": 10,
  "input": [
      {
          "text": "Calculate the compound interest on a principal of $10,000 invested for 5 years at an annual rate of 5%, compounded quarterly.",
          "label": "Simple task"
      },
      {
          "text": "分析使用CRISPR基因编辑技术在人类胚胎中的伦理影响。考虑潜在的医疗益处和长期社会后果。",
          "label": "Complex reasoning"
      },
      {
          "text": "AIが自意識を持つディストピアの未来を舞台にした短編小説を書いてください。人間とAIの関係や意識の本質をテーマに探求してください。",
          "label": "Creative writing"
      },
      {
          "text": "Erklären Sie die Unterschiede zwischen Merge-Sort und Quicksort-Algorithmen in Bezug auf Zeitkomplexität, Platzkomplexität und Leistung in der Praxis.",
          "label": "Complex reasoning"
      },
      {
          "text": "Write a poem about the beauty of nature and its healing power on the human soul.",
          "label": "Creative writing"
      },
      {
          "text": "Translate the following sentence into French: The quick brown fox jumps over the lazy dog.",
          "label": "Simple task"
      }
  ]
}
```

### Response
#### Success Response (200)
- **usage** (object) - Information about token usage.
  - **total_tokens** (integer) - The total number of tokens used.
- **data** (array) - An array of classification results.
  - **object** (string) - The type of object, always "classification".
  - **index** (integer) - The index of the classification result.
  - **prediction** (string) - The predicted label for the input.
  - **score** (float) - The confidence score for the prediction.
  - **predictions** (array) - An array of all possible label predictions and their scores.
    - **label** (string) - The label name.
    - **score** (float) - The confidence score for this label.

#### Response Example
```json
{
  "usage": {
    "total_tokens": 196
  },
  "data": [
    {
      "object": "classification",
      "index": 0,
      "prediction": "Simple task",
      "score": 0.35216382145881653,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.35216382145881653
        },
        {
          "label": "Complex reasoning",
          "score": 0.3412695527076721
        },
        {
          "label": "Creative writing",
          "score": 0.3065665662288666
        }
      ]
    },
    {
      "object": "classification",
      "index": 1,
      "prediction": "Complex reasoning",
      "score": 0.34310275316238403,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.32429200410842896
        },
        {
          "label": "Complex reasoning",
          "score": 0.34310275316238403
        },
        {
          "label": "Creative writing",
          "score": 0.3326052129268646
        }
      ]
    },
    {
      "object": "classification",
      "index": 2,
      "prediction": "Creative writing",
      "score": 0.3487184941768646,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.3321263790130615
        },
        {
          "label": "Complex reasoning",
          "score": 0.31915512681007385
        },
        {
          "label": "Creative writing",
          "score": 0.3487184941768646
        }
      ]
    },
    {
      "object": "classification",
      "index": 3,
      "prediction": "Complex reasoning"
    }
  ]
}
```
```

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

### GET /v1/classifiers

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/classifier-manage.html

Retrieves a list of all classifiers created by the user.

```APIDOC
## GET /v1/classifiers

### Description
List all classifiers you have created.

### Method
GET

### Endpoint
https://api.jina.ai/v1/classifiers

### Response
#### Success Response (200)
- **classifier_id** (string) - Unique identifier for the classifier
- **model_name** (string) - Name of the model used
- **labels** (array) - List of classification labels
- **access** (string) - Access level (e.g., private)
- **updated_number** (integer) - Number of updates
- **used_number** (integer) - Number of times used
- **created_at** (string) - Timestamp of creation
- **updated_at** (string) - Timestamp of last update
- **used_at** (string|null) - Timestamp of last usage
- **metadata** (object) - Additional metadata

#### Response Example
[
  {
    "classifier_id": "6db95bec-a2c4-4544-91de-11b863ba2bd9",
    "model_name": "jina-clip-v1",
    "labels": ["Food and Dining", "Nature and Outdoors"],
    "access": "private",
    "updated_number": 1,
    "used_number": 0,
    "created_at": "2024-10-30T12:51:51.241620+00:00",
    "updated_at": "2024-10-30T12:51:51.241620+00:00",
    "used_at": null,
    "metadata": {}
  }
]
```

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

### Requesting content via JavaScript fetch

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/r.reader.html

Perform a GET request to the Jina AI Reader API using the Fetch API.

```javascript
fetch('https://r.jina.ai/https://example.com', {
  method: 'GET',
  headers: {
    "Authorization": "Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq",
    "Accept": "application/json"
  },
})
```

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

### GET https://g.jina.ai

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Performs fact-checking and grounding on a provided query string.

```APIDOC
## GET https://g.jina.ai

### Description
Grounding API to fact-check a query string.

### Method
GET

### Endpoint
https://g.jina.ai

### Parameters
#### Query Parameters
- **search_parameters** (string) - Optional - Additional search parameters.
- **reference_urls** (string) - Optional - Comma-separated list of URLs for grounding references.
- **use_post** (boolean) - Optional - Use POST instead of GET.
- **json_response** (boolean) - Required - Indicates if the response will be in JSON format.
- **use_proxy** (boolean) - Optional - Whether to use a proxy server.
- **bypass_cache** (boolean) - Optional - To bypass the cache.
- **browser_locale** (string) - Optional - To control the browser locale.

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code.
- **status** (integer) - Custom status code.
- **data** (object) - Contains factuality score, result, reasoning, and references.
```

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

### Request Grounding via g.reader API

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Sends a GET request to the g.reader API to fact-check a query. Requires an authorization token and specific parameters for grounding configuration.

```python
endpoint = "https://g.jina.ai"  # The base URL to the grounding API (required)
query = "Jina AI was founded in 2020 in Berlin."  # Query string to be fact-checked (required)
headers = {
  "Accept": "application/json",  # Indicates that the response should be in JSON format (required)
  "Authorization": "Bearer <token here>"  # Authentication token to access the API (required)
}
params = {
  "search_parameters": "",  # Additional search parameters or headers (optional)
  "reference_urls": "",  # Comma-separated list of URLs for grounding references (optional)
  "use_post": False,  # Use POST instead of GET method, boolean value (optional)
  "json_response": True,  # Indicates if the response will be in JSON format, boolean value (required)
  "use_proxy": False,  # Whether to use a proxy server, boolean value (optional)
  "bypass_cache": False,  # To bypass the cache, boolean value (optional)
  "browser_locale": ""  # To control the browser locale, string value (optional)
}
response = requests.get(endpoint, params=params, headers=headers)  # Sends a GET request to the API
```

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

### Train Classifier with Text Data

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/classifier-train-text-image.html

Use this snippet to train a classifier with text-based training data. Ensure you provide a valid classifier ID and training examples with corresponding labels.

```bash
curl https://api.jina.ai/v1/train \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq" \
  -d @- <<EOFEOF
      {
        "classifier_id": "6db95bec-a2c4-4544-91de-11b863ba2bd9",
        "num_iters": 10,
        "input": [
            {
                "text": "A sleek smartphone with a high-resolution display and multiple camera lenses",
                "label": "Technology and Gadgets"
            },
            {
                "text": "Fresh sushi rolls served on a wooden board with wasabi and ginger",

```

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

### Fetch Meta-Prompt Content via Curl

Source: https://github.com/jina-ai/meta-prompt/blob/main/index.html

Use this command to fetch the specific version of the Meta-Prompt content programmatically. Ensure you replace {{CURL_COMMAND}} with the actual command.

```bash
curl {{CURL_COMMAND}}
```

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

### Pipe Meta-Prompt to LLM

Source: https://github.com/jina-ai/meta-prompt/blob/main/README.md

Use curl to fetch a specific version of the meta-prompt and pipe it directly into the llm CLI tool for processing.

```bash
curl docs.jina.ai/v1 | llm -s 'grab all sentences from Hacker News, embed them, and visualize the results in a 2D UMAP with matplotlib' -m claude-3.5-sonnet
```

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

### Access LLM via CLI

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/test-cases/docsqa.txt

Use this command to interact with a specified LLM model. Ensure the LLM is configured and accessible from your command line.

```shell
llm 'your prompt' -m claude-3.5-sonnet
```

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

### Requesting content via cURL

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/r.reader.html

Use this command to fetch URL content as JSON. Requires a valid Bearer token in the Authorization header.

```bash
curl 'https://r.jina.ai/https://example.com' \
	-H "Accept: application/json" \
	-H "Authorization: Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq"
```

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

### GET /v1/classifiers

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Retrieves a list of available classifiers with their metadata and usage statistics.

```APIDOC
## GET /v1/classifiers

### Description
Lists all classifiers accessible to the user, including details like model name, supported labels, and usage counts.

### Method
GET

### Endpoint
https://api.jina.ai/v1/classifiers

### Parameters
#### Query Parameters
- **param1** (string) - Optional - Example query parameter.
- **param2** (string) - Optional - Example query parameter.

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code.
- **status** (integer) - Internal status code.
- **data** (object) - Contains a list of classifier objects.

#### Response Example
{
  "code": 200,
  "status": 20000,
  "data": {
    "classifiers": [
      {
        "classifier_id": "string",
        "model_name": "string",
        "labels": ["label1"],
        "access": "public/private",
        "updated_number": 1,
        "used_number": 0,
        "created_at": "2024-10-30T12:51:51.241620+00:00"
      }
    ]
  }
}
```

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

### List Classifiers using cURL

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/classifier-manage.html

Use this command to retrieve a list of all classifiers associated with your account. Ensure your Authorization header is correctly set.

```bash
curl https://api.jina.ai/v1/classifiers \
	 -H "Content-Type: application/json" \
	 -H "Authorization: Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq"
```

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

### Classification API Response Formats

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Examples of successful and error response structures returned by the classification API.

```json
{
  "code": 200,
  "status": "success",
  "data": [
    {
      "object": "classification",
      "index": 0,
      "prediction": "Technology and Gadgets",
      "score": 0.30329811573028564,
      "predictions": [
        {"label": "Technology and Gadgets", "score": 0.30329811573028564},
        {"label": "Food and Dining", "score": 0.22840788960456848},
        {"label": "Nature and Outdoors", "score": 0.2357397824525833},
        {"label": "Urban and Architecture", "score": 0.23255419731140137}
      ]
    },
    ...
  ],
  "usage": {
    "total_tokens": 12065
  }
}
```

```json
{
  "code": 400,
  "status": "error",
  "message": "Invalid token provided"
}
```

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

### POST /v1/train

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Endpoint to initiate the training process for a classifier or model by providing input data and configuration parameters.

```APIDOC
## POST /v1/train

### Description
Initiates the training process for a specified classifier or model using provided input data.

### Method
POST

### Endpoint
https://api.jina.ai/v1/train

### Parameters
#### Request Body
- **classifier_id** (string) - Required - The ID of the classifier to train.
- **model** (string) - Required - Specifies the model to be used for training.
- **access** (string) - Required - Determines if the model is private or public.
- **num_iters** (integer) - Optional - The number of training iterations.
- **input** (array) - Required - Training data input containing text or image fields and labels.

### Request Example
{
  "classifier_id": "6db95bec-a2c4-4544-91de-11b863ba2bd9",
  "num_iters": 10,
  "input": [
    {
      "text": "A sleek smartphone",
      "label": "Technology"
    }
  ]
}

### Response
#### Success Response (200)
- **code** (integer) - HTTP status code.
- **status** (integer) - API specific status code.
- **data** (object) - Contains training results, usage statistics, and predictions.

#### Response Example
{
  "code": 200,
  "status": 20000,
  "data": {
    "usage": {
      "total_tokens": 120440
    }
  }
}
```

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

### POST /

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/segmenter.html

Performs text segmentation and tokenization on the provided content.

```APIDOC
## POST /

### Description
Segments the input text into semantically meaningful chunks and optionally returns token information.

### Method
POST

### Endpoint
https://segment.jina.ai/

### Parameters
#### Request Body
- **content** (string) - Required - The text content to be segmented.
- **return_tokens** (boolean) - Optional - Whether to return the tokens and their IDs.
- **return_chunks** (boolean) - Optional - Whether to return the segmented chunks.
- **max_chunk_length** (integer) - Optional - Maximum number of characters in each chunk.
- **head** (integer) - Optional - Return the first N tokens.
- **tail** (integer) - Optional - Return the last N tokens.
- **tokenizer** (string) - Optional - The tokenizer to use (e.g., cl100k_base).

### Request Example
{
  "content": "Jina AI: Your Search Foundation, Supercharged!",
  "return_tokens": true,
  "return_chunks": true,
  "max_chunk_length": 1000
}

### Response
#### Success Response (200)
- **num_tokens** (integer) - Total number of tokens.
- **num_chunks** (integer) - Total number of chunks.
- **chunk_positions** (array) - List of start and end indices for chunks.
- **chunks** (array) - List of segmented text strings.
- **tokens** (array) - List of tokens and their corresponding IDs.

#### Response Example
{
  "num_tokens": 78,
  "tokenizer": "cl100k_base",
  "num_chunks": 4,
  "chunk_positions": [[3, 55], [55, 93], [93, 110], [110, 135]],
  "tokens": [...]
}
```

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

### Embeddings API Response Structure

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/embeddings.html

This is an example of a successful response from the Embeddings API. It includes metadata about the request and the generated embeddings for each input text.

```json
{
  "model": "jina-embeddings-v3",
  "object": "list",
  "usage": {
    "total_tokens": 410,
    "prompt_tokens": 410
  },
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [
        0.0086874915,
        0.10442207,
        0.049341407,
        -0.09133959,
        0.035539553,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 1,
      "embedding": [
        -0.017281326,
        0.11304399,
        0.026589902,
        -0.05840733,
        0.038798418,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 2,
      "embedding": [
        -0.049376693,
        0.11805992,
        0.024820223,
        -0.07981897,
        0.010398907,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 3,
      "embedding": [
        0.059242655,
        0.043504987,
        0.021679124,
        -0.058159206,
        0.107634515,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 4,

```

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

### Segmenter API Response Example

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/segmenter.html

The response from the Segmenter API is a JSON object containing token and chunk information. It is not wrapped in a 'data' field.

```json
{ "num_tokens": 9735, "num_chunks": 62, "chunk_positions": [ [0, 4], [4, 401], [401, 866], [866, 1272], [1272, 1735], [1735, 2153], [2153, 2617], [2617, 3081], [3081, 3557], [3557, 4020], [4020, 4481], [4481, 4892], [4892, 5360] ], "chunks": [ "Australia/Lord_Howe is the weirdest timezone", "Classic 3D videogame shadow techniques", "Async Rust is not safe with io_uring", "Malibal and Coreboot Have a Disagreement", "2-adic numbering for binary tilings" ] }
```

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

### Read URL with JavaScript Fetch API

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/g.reader.html

This JavaScript snippet demonstrates how to fetch content from a URL using the Fetch API. It includes necessary headers for authorization and content acceptance.

```javascript
fetch('https://g.jina.ai/Jina%20AI%20was%20founded%20in%202020%20in%20Berlin.', {
      method: 'GET',
      headers: {
        "Authorization": "Bearer jina_74e70f6c60524b30811f57587baeece4EeJ603iFt-LCsemqEJHOJzLzgRnq",
        "Accept": "application/json"
      },
    })
```

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

### Jina AI Classifier API Response

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/classifier-classify-text.html

Example JSON response from the Jina AI Classifier API, showing classification predictions and confidence scores for each input.

```json
{
  "usage": {
    "total_tokens": 196
  },
  "data": [
    {
      "object": "classification",
      "index": 0,
      "prediction": "Simple task",
      "score": 0.35216382145881653,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.35216382145881653
        },
        {
          "label": "Complex reasoning",
          "score": 0.3412695527076721
        },
        {
          "label": "Creative writing",
          "score": 0.3065665662288666
        }
      ]
    },
    {
      "object": "classification",
      "index": 1,
      "prediction": "Complex reasoning",
      "score": 0.34310275316238403,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.32429200410842896
        },
        {
          "label": "Complex reasoning",
          "score": 0.34310275316238403
        },
        {
          "label": "Creative writing",
          "score": 0.3326052129268646
        }
      ]
    },
    {
      "object": "classification",
      "index": 2,
      "prediction": "Creative writing",
      "score": 0.3487184941768646,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.3321263790130615
        },
        {
          "label": "Complex reasoning",
          "score": 0.31915509700775146
        },
        {
          "label": "Creative writing",
          "score": 0.3487184941768646
        }
      ]
    },
    {
      "object": "classification",
      "index": 3,
      "prediction": "Complex reasoning",
      "score": 0.35207709670066833,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.34567955136299133
        },
        {
          "label": "Complex reasoning",
          "score": 0.35207709670066833
        },
        {
          "label": "Creative writing",
          "score": 0.30224335193634033
        }
      ]
    },
    {
      "object": "classification",
      "index": 4,
      "prediction": "Creative writing",
      "score": 0.3638903796672821,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.33112505078315735
        },
        {
          "label": "Complex reasoning",
          "score": 0.30498453974723816
        },
        {
          "label": "Creative writing",
          "score": 0.3638903796672821
        }
      ]
    },
    {
      "object": "classification",
      "index": 5,
      "prediction": "Simple task",
      "score": 0.3561534285545349,
      "predictions": [
        {
          "label": "Simple task",
          "score": 0.3561534285545349
        },
        {
          "label": "Complex reasoning",
          "score": 0.323111891746521
        },
        {
          "label": "Creative writing",
          "score": 0.3207346975803375
        }
      ]
    }
  ]
}
```

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

### Train a Model Request

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/index-gen.txt

Sends a POST request to train a model with specific parameters and input data.

```python
endpoint = "https://api.jina.ai/v1/train" # Endpoint URL for training a model

# Header specifying the content type of the request body
headers = {
    "Content-Type": "application/json", # Required: Indicates the media type of the resource
    "Authorization": "Bearer <token here>", # Required: Authentication token to validate the request
    "Accept": "application/json" # Optional: Indicates that the client expects a JSON response
}

# Data payload containing the training parameters and input data
data = {
    "model": "jina-embeddings-v3", # Required: Specifies the model to be used for training
    "access": "private", # Required: Determines if the model is publicly accessible or private
    "num_iters": 10, # Optional: Number of training iterations, defaults to a value that optimizes performance
    "input": [ # Required: Input data for training the model
        {
            "text": "Calculate the compound interest on a principal of $10,000 invested for 5 years at an annual rate of 5%, compounded quarterly.",
            "label": "Simple task"
        },
        # Additional input data objects...
    ]
}

# Making a POST request to the endpoint with headers and data
response = requests.post(endpoint, json=data, headers=headers)
```

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

### Default Meta-Prompt Access via Curl

Source: https://github.com/jina-ai/meta-prompt/blob/main/index.html

This command demonstrates how to access the default version of the Meta-Prompt content. It's a simple curl command for direct text retrieval.

```bash
curl docs.jina.ai
```

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

### Jina Embeddings API Response Structure

Source: https://github.com/jina-ai/meta-prompt/blob/main/experiment/html-snippets/embeddings.html

This is an example of a successful response from the Jina embeddings API. It includes the model used, usage statistics, and a list of embeddings, each with an index and the embedding vector.

```json
{
  "model": "jina-clip-v1",
  "object": "list",
  "usage": {
    "total_tokens": 5033,
    "prompt_tokens": 5033
  },
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [
        -0.008161531,
        0.0040345504,
        -0.037822176,
        -0.012172974,
        -0.06706689,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 1,
      "embedding": [
        -0.032783166,
        0.005709158,
        -0.05175277,
        0.018007655,
        -0.026261164,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 2,
      "embedding": [
        -0.04354628,
        0.014823603,
        -0.0219809,
        0.013545886,
        -0.08526429,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 3,
      "embedding": [
        -0.021557946,
        0.010585106,
        0.014772621,
        -0.062192347,
        0.01447019,
        ...
      ]
    },
    {
      "object": "embedding",
      "index": 4,
      "embedding": [
        -0.05983584,
        0.012545062,
        0.0022877986,
        0.03144552,
        0.0013024239,
        ...
      ]
    }
  ]
}
```