AI Humanizer API That Scales With Your Client Workload
One API to humanize every client article and verify it reads human before it ships. Word-based pricing that starts around $0.17 per 1,000 words and scales from 125,000 to 400,000 words a month.
Word-based pricing from $0.17 per 1K words. Batch-ready. 40+ languages.
Built for content at scale
One REST endpoint your team loops over to humanize every client article before it goes live.
Pricing tied to client words, not per-call fees
You are billed by input words on POST /v1/humanize, so cost tracks the articles you actually process. Standard is 125,000 words a month and Premium is 400,000, with an effective rate starting around $0.17 per 1,000 words.
Built to loop through batches
Premium handles 120 requests per minute and Standard 40, with up to 2,000 input words per request. Queue a client's article set, loop the calls, and let your pipeline run unattended.
Every article verified human before publish
POST /v1/detect returns ai_probability and predicted_class so you can gate publishing. Nothing ships until it passes your AI-detection check, the same scoring behind /ai-detector.
Top-ups that absorb client spikes
When a launch month runs hot, add 125,000-word top-up packs that never expire. No overage surprises, no throttled delivery, no renegotiating a plan mid-sprint.
Pricing that scales with your client roster
Point your pipeline at one loop: pull each client draft, POST it to /v1/humanize, score the result with /v1/detect, and publish only what reads human. With up to 2,000 input words per request and 120 requests per minute on Premium, a full client batch runs unattended and word-based billing keeps cost tied to the articles you actually process.
Boutique
~120,000 words / month
Standard at $29 a month covers 125,000 words, enough for a small client roster's monthly output.
Growing
~400,000 words / month
Premium at $69 a month covers 400,000 words plus 3 scored variations, sized for a multi-client content calendar.
High-volume
~525,000+ words / month
Premium's 400,000 words plus 125,000-word top-up packs that never expire, so programmatic and launch-month spikes are absorbed without overage.
Estimates assume roughly 1,200 words per article. Words are billed on input text.
Throughput your delivery schedule can rely on
Predictable per-word cost, fast responses, and one endpoint to batch against.
Pull drafts from your content pipeline
Take the AI-generated drafts coming out of your writers or generation step and queue them for humanizing across all your client accounts.
Humanize in a batch loop
Loop each draft through POST /v1/humanize with a tone like blog or professional and let auto-detected language handle multilingual clients across 40+ languages.
Score, gate, then publish
Run POST /v1/detect on each result and publish only when predicted_class reads human, so every piece clears your AI-detection check before it goes live.
Loop your whole batch through one endpoint
Point your pipeline at one loop: pull each client draft, POST it to /v1/humanize, score the result with /v1/detect, and publish only what reads human. With up to 2,000 input words per request and 120 requests per minute on Premium, a full client batch runs unattended and word-based billing keeps cost tied to the articles you actually process.
curl -X POST https://api.writehuman.ai/v1/humanize \
-H "Authorization: Bearer wh_live_your_key_here" \
-H "Content-Type: application/json" \
-d '{
"text": "Your client draft goes here. Up to 2000 words per request.",
"tone": "blog",
"language": "en"
}'Simple, word-based pricing
Pay for the words you send. Allowances reset every billing cycle, and top-up packs never expire.
Standard
$29 / mo- 125,000 words / month
- 40 requests / min
- Single humanized result per request
Premium
$69 / mo- 400,000 words / month
- 120 requests / min
- 3 variations with human-likeness scores
Frequently asked questions
Will pricing blow up when we scale to agency volume?
No. Plans are priced by input words and reset each cycle: Standard is $29 a month for 125,000 words and Premium is $69 a month for 400,000 words, an effective rate starting around $0.17 per 1,000 words. When a month runs hot, add 125,000-word top-up packs that never expire, so you absorb client spikes without overage surprises.
Can the API handle batch or programmatic throughput?
Yes. Premium allows 120 requests per minute and Standard 40, with up to 2,000 input words per request and an average response around 20 seconds. The endpoints are built for a loop-friendly batch pattern, so you can run a client's full article set through your pipeline unattended. See /api/docs for the integration details.
How do we know an article reads human before it goes live?
Call POST /v1/detect on each humanized article. It returns ai_probability from 0 to 1 and a predicted_class of ai, human, or mixed, with an optional per-sentence breakdown. Use that as a publish gate so nothing ships until it passes your AI-detection check. It is the same scoring behind /ai-detector.
Does it support multilingual clients?
Yes. The humanize endpoint covers 40-plus languages and auto-detects the input, so you can run English, Spanish, German, and more through the same batch loop. You can also pass an explicit language value when you want to be precise about a client's locale.
How do we manage keys across an agency team?
Authenticate with a Bearer token using the wh_live_ key prefix. You can hold up to 5 keys per account and revoke any of them at any time, which makes it straightforward to separate keys by client, project, or environment and rotate them when staff change.
What tones fit different client content?
Pass a tone on each humanize call to match the client brief: professional, academic, blog, casual, creative, scientific, or technical. That lets one pipeline serve a B2B SaaS blog and a consumer lifestyle site without retraining your process.
Plug WriteHuman into your content pipeline
Get a wh_live_ key, loop your client drafts through one humanize and detect pass, and ship only the articles that read human. Start on Standard at 125,000 words, scale to Premium at 400,000, and top up by 125,000 words whenever a client launch spikes. See plans at /api#pricing or read the integration guide at /api/docs.