Skip to main content

Summarization

Extract key sentences and segments from text using prompted extractive summarization.


Endpoint​

POST https://api.platform.knowledgator.com/inference/extract/summary/


Headers​

NameTypeDescription
KnowledgatorAPI-KeyStringYour API key
Content-TypeStringapplication/json
KnowledgatorAPI-Model-IDStringOptional model ID for using a specific fine-tuned model

Request Body​

{
"text": "Your input document here.",
"prompt": "Extract sentences that summarize the main findings.",
"threshold": 0.5
}

Parameters​

  • text (string, required): Input document to summarize.
  • prompt (string, optional): Task instruction to guide what to extract. Examples:
    • "Extract sentences that summarize the main findings."
    • "Extract key parts describing risks and limitations."
    • "Identify the most important conclusions."
  • threshold (float, optional): Filtering threshold for summary spans. Default: 0.5

Response​

{
"outputs": [
{
"text": "The study found significant improvements in accuracy.",
"start": 0,
"end": 56,
"score": 0.91
},
{
"text": "Further research is needed to validate these results.",
"start": 128,
"end": 181,
"score": 0.87
}
]
}

Response Fields​

  • outputs: Array of extracted summary spans
    • text: Extracted summary span (sentence or segment)
    • start: Character offset (0-based, inclusive)
    • end: Character offset (0-based, exclusive)
    • score: Model confidence that this span is part of the summary

Error Codes​

Status CodeDescription
200OK – Request successful.
400Bad Request – Invalid or incomplete input.
401Unauthorized – Invalid API key.
404Not Found – Invalid endpoint.
413Payload Too Large – Exceeds size limit.
500Internal Server Error – Server-side issue.

Example Usage​

SDK Usage​

💡 Dedicated SDK Guides: See Python SDK | TypeScript SDK

TypeScript/JavaScript​

import { createClient } from '@knowledgator/typescript-sdk';

const kg = createClient('YOUR_API_KEY');

const result = await kg.extract.summary({
text: `
The study examined the effects of climate change on coastal ecosystems.
Researchers found significant coral bleaching in tropical regions.
Temperature increases of 2°C led to 40% habitat loss.
The team recommends immediate conservation efforts.
Further research is needed to validate these results across different regions.
`,
prompt: "Extract the main findings and recommendations.",
threshold: 0.5,
adapter_id: "your-lora-adapter-id" // Optional LoRA adapter
});

for (const span of result.outputs) {
console.log(`[${span.score.toFixed(2)}] ${span.text}`);
}

Python​

from knowledgator import Knowledgator

kg = Knowledgator(api_key="YOUR_API_KEY")

result = kg.extract.summary(
text="""
The study examined the effects of climate change on coastal ecosystems.
Researchers found significant coral bleaching in tropical regions.
Temperature increases of 2°C led to 40% habitat loss.
The team recommends immediate conservation efforts.
Further research is needed to validate these results across different regions.
""",
prompt="Extract the main findings and recommendations.",
threshold=0.5,
adapter_id="your-lora-adapter-id" # Optional LoRA adapter
)

for summary_span in result.summary.value:
print(f"[{summary_span.score:.2f}] {summary_span.text}")

Expected SDK Output​

[0.91] Researchers found significant coral bleaching in tropical regions.
[0.86] Temperature increases of 2°C led to 40% habitat loss.
[0.84] The team recommends immediate conservation efforts.

Python (requests)​

import requests

url = "https://api.platform.knowledgator.com/inference/extract/summary/"
headers = {
"Content-Type": "application/json",
"KnowledgatorAPI-Key": "YOUR_API_KEY",
"KnowledgatorAPI-Model-ID": "your-model-id" # Optional
}
payload = {
"text": """
The study examined the effects of climate change on coastal ecosystems.
Researchers found significant coral bleaching in tropical regions.
Temperature increases of 2°C led to 40% habitat loss.
The team recommends immediate conservation efforts.
Further research is needed to validate these results across different regions.
""",
"prompt": "Extract the main findings and recommendations.",
"threshold": 0.5
}

response = requests.post(url, json=payload, headers=headers)
print(response.json())

cURL​

curl -X POST https://api.platform.knowledgator.com/inference/extract/summary/ \
-H "Content-Type: application/json" \
-H "KnowledgatorAPI-Key: YOUR_API_KEY" \
-H "KnowledgatorAPI-Model-ID: your-model-id" \
-d '{
"text": "The study examined the effects of climate change on coastal ecosystems. Researchers found significant coral bleaching in tropical regions. Temperature increases of 2°C led to 40% habitat loss. The team recommends immediate conservation efforts. Further research is needed to validate these results across different regions.",
"prompt": "Extract the main findings and recommendations.",
"threshold": 0.5
}'

Expected Response (cURL & requests)​

{
"outputs": [
{
"span": "Researchers found significant coral bleaching in tropical regions.",
"start": 82,
"end": 149,
"score": 0.91
},
{
"span": "Temperature increases of 2°C led to 40% habitat loss.",
"start": 154,
"end": 207,
"score": 0.86
},
{
"span": "The team recommends immediate conservation efforts.",
"start": 212,
"end": 263,
"score": 0.84
}
]
}

Additional Information​

  • Extractive vs Abstractive: This is extractive summarization - it selects existing sentences/spans rather than generating new text
  • Prompt Flexibility: The prompt can guide what aspects to focus on (findings, risks, conclusions, etc.)
  • Use Cases:
    • Document summarization
    • Key point extraction
    • Meeting notes condensation
    • Research paper highlights
    • Legal document analysis