REST API / text rewriting

AI Humanizer API for editable, natural rewrites

Send a draft, choose a humanizer model, and receive a new version inside your own workflow. The response stays easy to edit, compare, and approve before it goes anywhere public.

Create an API key Read the docs

Use the key from a private server environment.

cURL request

curl --request POST \
  --url https://api.detecting-ai.com/api/humanize/ \
  --header "Content-Type: application/json" \
  --header "Idempotency-Key: humanizer-review-001" \
  --header "X-API-Key: YOUR_API_KEY" \
  --data '{
    "text": "Paste the text you want to rewrite.",
    "model": "cognia"
  }'

POST requests to /api/humanize/

X-API-Key authentication

Client Idempotency-Key for recoverable retries

Four supported models

Word usage in every response

A rewriting step you control

Keep the source, the rewrite, and the final edit in one flow

A useful humanizer integration does not hide what changed. Store the original, request a rewrite, and give an editor a clear place to compare the two versions before approving the result.

Simple JSON input

Send the draft and one supported model name. The response returns plain text plus the model and processed word count.

Model choice stays explicit

Select cognia, lexi, cognia_v2, or huma_v2 based on the behavior your team has tested for its own content.

Editable by design

Put the returned text back into an editor, revision queue, or content system instead of treating it as a finished artifact.

Request contract

Send the draft and name the model

Both fields are required by the REST endpoint. Keep model selection in configuration so it can be tested and changed without rewriting the integration.

Field	Type	Required	What it does
text	string	Yes	The supplied text you want the service to rewrite.
model	string	Yes	One of cognia, lexi, cognia_v2, or huma_v2.

Three fields make the result easy to track

Use humanized_text as the editable result. Record model_used and words_processed beside it so support, billing, and content teams can understand how the request ran.

Example response

{
  "humanized_text": "The rewritten text is returned here.",
  "words_processed": 8,
  "model_used": "cognia"
}

Integration flow

Treat humanization as an editing pass

The strongest workflow keeps the original visible and asks a person to approve the new version.

1
Protect the API key
Store the key on your server or in a secret manager. Do not place it in public browser code or a shared document.
2
Send text with a tested model
Choose one supported model, persist a unique Idempotency-Key before sending, and reuse that key only for an exact retry.
3
Review and approve
Check the rewritten text for accuracy, meaning, tone, and required terminology before saving or publishing it.

Good fits

A practical fit for products that already have an editor

Humanization is most useful when a draft already moves through review. The API adds a rewrite option without replacing the people responsible for the final words.

A rewrite can change meaning

Names, quotations, figures, legal language, and technical terms deserve a careful check. Keep those details visible to the reviewer.

Content management

Offer a rewrite action inside a CMS and return the result to the same draft for comparison and approval.

Support operations

Revise approved source material into clearer help text, then send it through the normal fact and policy review.

Editorial queues

Add a rewrite candidate beside the original so an editor can keep, combine, or reject individual changes.

Private writing tools

Call the service from a trusted backend and expose only the returned text your signed-in user needs.

API FAQ

Practical answers before you ship

Start with a server-side request, log the response shape, then add the result to your existing review flow.

What endpoint does the AI Humanizer API use?+

Send a POST request to https://api.detecting-ai.com/api/humanize/. Use JSON, pass your key in X-API-Key, and persist a stable Idempotency-Key for every logical operation so retries are recoverable.

Which humanizer models are available?+

The supported model values are cognia, lexi, cognia_v2, and huma_v2. The REST API expects the model field on every request.

What does the response contain?+

A successful response contains humanized_text, words_processed, and model_used. The rewritten text remains editable in your product.

Should the API output be published without review?+

No. Review the result for meaning, facts, tone, names, links, and any language that must stay exact. A rewrite can improve flow while still needing an editor.

How is humanizer usage counted?+

Usage is based on the input word count and comes from the humanizer balance attached to the API subscription. Successful responses report words_processed.

Build the first request

Test the rewrite on text you know well

Start with one model, compare the result against the source, and decide which checks your editors need before the feature ships.

Open API docs Compare pricing