High similarity
Matched text is linked to its source and character range.
Scan text against web sources and receive a plagiarism score, matched source URLs, highlighted character ranges, and review-ready statistics.
The Eiffel Tower is a wrought-iron lattice tower on the Champ de Mars in Paris, France. It is named after the engineer Gustave Eiffel, whose company designed and built the tower from 1887 to 1889.
Matched text is linked to its source and character range.
en.wikipedia.org/wiki/Eiffel_Tower
curl -X POST https://developer-portal.walterwrites.ai/api/plagiarism/scan/
-H "Content-Type: application/json"
-H "X-API-Key: YOUR_API_KEY"
-d '{
"content": "The Eiffel Tower is a wrought-iron lattice tower on the Champ de Mars in Paris, France. It is named after the engineer Gustave Eiffel, whose company designed and built the tower from 1887 to 1889."
}'"status": "success",
"plagiarism_score": 0.99,
"verdict": "severe_plagiarism",
"sources": [
{
"url": "https://en.wikipedia.org/wiki/Eiffel_Tower",
"similarity_score": 0.97,
"matched_word_count": 34
}
],
"stats": { "word_count": 37, "sources_checked": 24, "matched_word_count": 34 },
"credits_remaining": 1998A synchronous scan with matched sources, stable character ranges, and explicit similarity data for human review.
Return source titles, domains, URLs, similarity scores, and matched-word counts.
Map every returned start and end offset back to the text exactly as submitted.
Use documented score bands as workflow labels, then keep the final decision with a reviewer.
Receive word, character, sentence, source-count, and processing-time details.
Use an idempotency key to avoid duplicate scans and duplicate charges during retries.
Each completed scan costs 2 credits regardless of text length.
Use one developer portal and scoped API keys for humanization, AI text detection, grammar correction, plagiarism scanning, and AI image detection.


Analyze text for AI-generated content with confidence scores and sentence-level details.


Transform AI-generated text into natural, human-like content through a documented REST endpoint.


Audio detection is planned for a future Walter API release.


Scan text against web sources and return a score, matched URLs, and highlighted character ranges.


Correct grammar, spelling, and punctuation and return corrected text plus a complete word-level diff.


Upload an image and receive a real, fake, or inpainting verdict with confidence and per-class probabilities.
Each plagiarism scan costs a flat 2 credits regardless of text length. Failed scans are refunded automatically.
Need enterprise pricing?
Talk to us about higher-volume workloads and enterprise deployment requirements.
Documented answers about the endpoint, response, limits, billing, and safe production use.
Yes. Send a POST request to /api/plagiarism/scan/ with an API key that has the plagiarism_checker scope.
It returns a 0 to 1 plagiarism score, a verdict, a summary, matched sources, highlighted character ranges, scan statistics, execution time, and the remaining credit balance.
No. The API reports source similarity for review. Quotations, citations, templates, common phrasing, and licensed material can create legitimate overlap, so a person should inspect the matched passage and source.
Text must contain at least 50 characters and may contain up to 25,000 words. Word counting is script-aware.
Every completed scan costs a flat 2 credits regardless of text length. Failed scans are refunded automatically.
Send an Idempotency-Key header. Reusing the same key with the same text within five minutes returns the original response without another scan or charge.
Create an API key with the required scope, follow the documented request format, and keep the returned evidence in your existing review workflow.
From the blog
Guides, comparisons and practical advice from Walter.
View all