| What it does | AI humanizer for multilingual writing with best-in-class Japanese support
| Help docs that write themselves from your code
|
|---|
| Stage | Startup
| Not given
|
|---|
| Founded | 2026
| 2023
|
|---|
| Based in | HK
| Not given
|
|---|
| Pricing model | Freemium
| Freemium
|
|---|
| Pricing | Free: 500 words/day humanize + unlimited detector (no credit card required); Basic: $9.99/month for 50,000 words/month + 30-day history + priority email support; Pro: $19.99/month for 250,000 words/month + academic mode + 90-day history + priority API queue. Student discount: 50% off annual plans wi
| Free: $0 forever. Starter: $39 per month ($468 billed yearly). Growth: $99 per month ($1,188 billed yearly). Scale: $279 per month ($3,348 billed yearly). All paid plans include a 14-day free trial of Growth. Founding customers: 50% off first year on paid plans.
|
|---|
| Key features | - AI text humanizer with casual, academic, and business modes
- Free AI content detector with ai_score, perplexity, and burstiness metrics
- Morphology-aware Japanese rewriting with conjugation and particle control
- Citation and technical term preservation in academic mode
- Three parallel rewrites scored by separate detector model
- Five-step intensity slider for faithfulness vs. AI-fingerprint tradeoff
- History and re-rewrite capability (30 or 90 days depending on plan)
- Second-pass academic polish with Claude 3.5 Haiku on Pro plan
- Support for English, Japanese, Korean, German, French, Spanish, Portuguese, and Chinese
- Tested against GPTZero, Turnitin, Originality.ai, and Copyleaks
| - Drafts help centre articles from code
- Flags stale articles on merged pull requests
- Generates changelog from merged PRs
- Publishes API reference without OpenAPI spec
- AI answers on help centre
- Imports existing documentation
- Integrates with Claude and Cursor via MCP
- Links claims back to source code
- Draft review before publishing
|
|---|
| What makes it different | - Best-in-class Japanese support with morphology-aware morpheme-level tuning, unlike generic humanizers
- Separate AI detector scores rewrites to guarantee the most natural result
- Preserves academic citations, figures, tables, and technical terminology verbatim
- Five-step granular intensity control to trade off faithfulness versus AI-fingerprint removal
- Independent developer with transparent privacy: no model training on user text, fixed deletion schedule
| - Writes documentation directly from your code, not manual files
- Keeps docs accurate by flagging stale articles on each merged PR
- Captures real screenshots of your app rather than generic mockups
- Generates API reference automatically without an OpenAPI spec
- Works with apps built using no-code tools like Lovable, Bolt or v0
|
| Community votes | 0 | 0 |
| DR (Domain Rating by Ahrefs) | 4 ● no change since last check | 37 ● no change since last check |
| Trust Flow (Majestic) | 11 | 9 |
| Citation Flow (Majestic) | 14 | 37 |
| Referring domains (Majestic) | 46 | 603 |