Comments
>log in to commentHi, I'm Tomer, co-founder of Unprofane. Try a disguised insult in our playground, then try a harmless name like Scunthorpe. The result shows what the filter caught and why. If you're adding usernames or chat to an app, free test keys let you try it before paying. Our launch offer takes 50% off your plan's monthly price for the first three months when you start by January 31, 2027; code LAUNCH50. Tell us what your current filter gets wrong at hello@unprofane.com.
- Runs profanity rules at the edge, only calls model when needed
- Public Scunthorpe benchmark with transparent results
- Supports 17 languages with custom word lists
- Free test keys, paid plans start at $9/month
Unprofane is an API that detects and censors profanity, toxic language and disguised abuse in user-generated text.
- for
- Developers building apps with user-generated content such as chat, usernames or custom prints.
- pricing
- paid · free trial
Key features
- Disguise detection — Catches leetspeak, homoglyphs, zero-width characters and other obfuscation techniques.
- Multi-language support — Filters profanity in 17 languages with language-specific lexicons.
- Contextual policies — Provides preset policies (e.g., teen) for different content types.
- Live chat integration — Filters messages in real-time, returning censored text for display.
- Username validation — Checks usernames at sign-up, handling swapped letters and staff name exceptions.
- Custom text moderation — Reviews custom designs or prints before they are processed.
- Privacy by design — No request text is stored; Edge-only projects keep data off the cloud.
- SDKs & recipes — Ready-to-use integrations for Supabase, Clerk, Firebase, Auth0, Unity, Godot, Node.js and Python.
Use cases
- Filter profanity in in-game chat so players see censored messages instantly.
- Validate usernames during registration to block abusive or disguised names.
- Moderate custom merchandise text to prevent hateful designs from being printed.
- Screen user reviews or comments for toxicity before publishing.
- Replace Google Perspective API with a drop-in endpoint for existing integrations.
Unprofane pricing
- Hobby$9 a month/month25,000 checks · 1 project · Free test keys (100 checks/day)
- Starter$29 a month/month100,000 checks · 3 projects · Free test keys (100 checks/day)
- Growth$249 a month/month1,000,000 checks · 10 projects · Engine pinning and CSV replay
- Scale$999 a month/month5,000,000 checks · Unlimited projects · Priority support
- Researchfree250,000 checks/month for 12 months by application
Unprofane vs alternatives
 Unprofane |  Postman |  Insomnia |  Meilisearch |  Twilio | |
|---|---|---|---|---|---|
| Best for | Detecting leetspeak and disguised profanity | API design and testing | API debugging | Full-text search | Messaging infrastructure |
| Pricing | Paid | Subscription | Free | Free | Subscription |
| DevHunt upvotes | 1 | 0 | 0 | 0 | 0 |
| Launched | Oct 2026 | — | — | — | — |
- Unprofane vs Postman: General API testing platform, not a profanity filter
- Unprofane vs Insomnia: Open-source API client, no content moderation features
- Unprofane vs Meilisearch: Search engine API, unrelated to text moderation
- Unprofane vs Twilio: Communication APIs (SMS/voice), not profanity detection
Unprofane FAQ
Is there a free plan?+
Test keys are free and need no card, but live keys require a paid plan starting at $9/month.
What counts as a check?+
Each text sent to the check endpoint, batch endpoint item, Perspective endpoint or MCP server counts as one check.
Can I limit my bill?+
Yes, you can set a spend cap; once reached, requests return a quota-exceeded error.
Does Unprofane store user text?+
No, request text is never stored in a database, cache, log or trace.
How does it compare to Google Perspective API?+
In public tests Unprofane caught 89% of toxic text versus 58.4% for Perspective, with similar false-alarm rates.
Summarized by DevHunt from unprofane.com · Oct 14, 2026. Details may change; check the official site.
