Inference API

The Inference API allows you to submit text and receive an AI likelihood score.

Tip

Use the batch endpoint to maximize throughput when classifying multiple documents.

Note

text.api.pangramlabs.com and text-batch.api.pangramlabs.com only use the first ~400 words of the input text when making its prediction.

For accurate predictions on longer text with mixed human and AI content, break your document into chunks of ~400 words or use the sliding window API.

POST https://text.api.pangramlabs.com

Request JSON Object:

text (string) – The input text to classify.

Response JSON Object:

ai_likelihood (float) – The classification of the text, on a scale from 0.0 (human) to 1.0 (AI).
text (string) – The classified text.
prediction (string) – A string representing the classification.

Request Headers

{
  "Content-Type": "application/json",
  "x-api-key": "<api-key>"
}

Request Body

{
  "text": "<text>"
}

Example Request

POST https://text.api.pangramlabs.com HTTP/1.1
Content-Type: application/json
x-api-key: your_api_key_here

{
  "text": "The text to analyze"
}

Example Response

{
  "text": "The text to analyze",
  "prediction": "Likely AI",
  "ai_likelihood": 0.92
}

POST https://text-batch.api.pangramlabs.com

Request JSON Object:

text (array) – An array of input texts to classify.

Response JSON Object:

responses (array) – The classification results as a list, each item containing “text”, “ai_likelihood”, and “prediction”. Each item in the array is the same as a response from a single text prediction.

Batch Inference

Request Headers

{
  "Content-Type": "application/json",
  "x-api-key": "<api-key>"
}

Request Body

{
  "text": ["<text1>", "<text2>", "..."]
}

Example Request

POST https://text-batch.api.pangramlabs.com HTTP/1.1
Content-Type: application/json
x-api-key: your_api_key_here

{
  "text": ["The first text to analyze", "The second text to analyze"]
}

Example Response

{
  "responses": [
    {
      "text": "The first text to analyze",
      "prediction": "Likely AI",
      "ai_likelihood": 0.92
    },
    {
      "text": "The second text to analyze",
      "prediction": "Possibly AI",
      "ai_likelihood": 0.58
    }
  ]
}

POST https://text-sliding.api.pangramlabs.com

Request JSON Object:

text (string) – The input text to segment into windows and classify.

Response JSON Object:

text (string) – The classified text.
ai_likelihood (float) – The classification of the text, on a scale from 0.0 (human) to 1.0 (AI).
max_ai_likelihood (float) – The maximum AI likelihood score among all windows.
avg_ai_likelihood (float) – The average AI likelihood score among all windows.
prediction (string) – A string representing the classification.
fraction_ai_content (float) – The fraction of windows that are classified as AI.
windows (array) – A list of windows and their individual classifications. Each object in the array is the response from a single text prediction.

Request Headers

{
  "Content-Type": "application/json",
  "x-api-key": "<api-key>"
}

Request Body

{
  "text": "<text>"
}

Example Request

POST https://text.api.pangramlabs.com HTTP/1.1
Content-Type: application/json
x-api-key: your_api_key_here

{
  "text": "Extremely long text."
}

Example Response

{
  "text": "Extremely long text.",
  "prediction": "Highly likely AI",
  "ai_likelihood": 1.0,
  "max_ai_likelihood": 1.0,
  "avg_ai_likelihood": 0.6,
  "fraction_ai_content": 0.5,
  "windows": [
    {
      "text": "Extremely long",
      "ai_likelihood": 1.0,
      "prediction": "Highly likely AI"
    },
    {
      "text": "long text.",
      "ai_likelihood": 0.1,
      "prediction": "Unlikely AI"
    }
  ]
}

Errors

The API may return the following error codes:

400 Bad Request - If the request body is not properly formatted.
401 Unauthorized - If the x-api-key is missing, invalid, or does not have enough credits to process the request.
500 Internal Server Error - If there is an error processing the request.

Please reach out at support@pangram.com if you are running into errors with your requests.