summarize_text
Summarize long documents into key points using extractive and abstractive methods. Ideal for articles, reports, and transcripts.
Instructions
Summarize a long document or passage into key points.
Uses extractive and abstractive summarisation to distil the most important
information from the input. Suitable for news articles, reports, research
papers, meeting transcripts, and support tickets.
Args:
text: Text to summarise. Works best with 200+ words of input.
Handles articles, reports, and multi-paragraph passages.
max_sentences: Target number of sentences in the output summary
(default 3, range 1–10). The model may produce fewer
if the source is already concise.
Returns:
summary: The generated summary text.
sentences: Number of sentences in the summary.
compression_ratio: Input length divided by summary length.
key_points: Bullet-point list of the main points (may be null
for very short inputs).
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes | ||
| max_sentences | No |