Text Summarizer — condense long text blocks

Text Summarization API

What it does

Text Summarizer turns a single block of text into a shorter summary. Send text in the body.text field and get back the generated summary along with model, input_length, summary_length, and compression_ratio.

Use it when you need to condense long articles, support tickets, research notes, or meeting transcripts into something faster to review. The request accepts optional controls for model (bart or pegasus), num_beams, max_length, and min_length, so you can tune the tradeoff between brevity and detail.

The response is structured for downstream automation. input_length and summary_length help you measure how aggressively the text was shortened, while compression_ratio gives you a simple signal for reporting or quality checks. Keep the input between 50 and 16,384 characters and use the summary in search, classification, or human review workflows.

Quickstart

Summarize a piece of text by sending it in a JSON request body.

curl -X POST "http://localhost:8080/dosvak/summarize-text-api" \
  -H "apy-token: $APY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Artificial intelligence is transforming how teams work by automating repetitive tasks, improving search, and helping people draft content faster. Many organizations are adopting AI tools to increase productivity while keeping humans in the loop for review and decision-making."
  }'

What you'll get back

Returns a JSON object with model as a string, summary as a string, input_length as an integer, summary_length as an integer, and compression_ratio as a number.

{
  "model": "bart",
  "summary": "AI is helping teams work faster by automating tasks, improving search, and assisting with drafting, while humans still review and decide.",
  "input_length": 286,
  "summary_length": 24,
  "compression_ratio": 11.92
}

About this endpoint

What it does

Summarizes a single text block by sending the input text to the selected summarization model and returning the generated summary along with summary metadata.

Request Body

Parameter	Type	Mandatory	Description
text	String	Yes	The text to summarize. Must be between 50 and 16384 characters long.
model	ENUM	No	Summarization model to use. Allowed values: `bart`, `pegasus`. Default: `bart`.
num_beams	Integer	No	Beam search count. Minimum: 1, maximum: 8. Default: 4.
max_length	Integer	No	Maximum length of the generated summary. Minimum: 30, maximum: 512. Default: 200.
min_length	Integer	No	Minimum length of the generated summary. Minimum: 10, maximum: 200. Default: 40.

Response

Returns a JSON object with model and summary string fields, plus input_length and summary_length integer fields and a compression_ratio number field.

Parameter	Type	Mandatory	Description
model	String	No	The model used to generate the summary.
summary	String	No	The generated summary text.
input_length	Integer	No	Length of the input text.
summary_length	Integer	No	Length of the generated summary.
compression_ratio	Number	No	Compression ratio of the summary relative to the input.

Body

Name

Type

Description

bodyREQUIRED

object

—

Quickstart

What you'll get back

About this endpoint

What it does

Request Body

Response

Body

Errors any endpoint can return