| What it does | On-device moderation that flags hateful, abusive and threatening text in 23 European languages, before the message is sent.
| Take your AI support bot from 4/10 to 10/10
|
|---|
| Stage | MVP
| MVP
|
|---|
| Founded | Not given
| 2026
|
|---|
| Based in | NL
| Not given
|
|---|
| Pricing model | Freemium
| Free trial
|
|---|
| Pricing | Free up to 100k monthly active devices with unlimited inference. Custom licenses available.
| Free during beta. After beta: $29/month (your own AI keys, unlimited runs) or $59/month (AI included, 50 runs/month).
|
|---|
| Key features | - Three separate scores: hateful, abusive, threat
- Ten target signals under the hate score: race, color, religion, descent, national or ethnic origin, sexual orientation, gender, disability, age, other
- Runs fully on device with no server calls
- Supports 23 European languages from one model file
- 80MB model size optimized for iOS, Android, and browser
- 0.847 macro F1 accuracy on Multilingual HateCheck
- Warn before message is sent
- Comment triage without infrastructure
| - Bot Score: rates bot performance across five dimensions (rules, facts, solutions, tone, trick resistance)
- Break my bot: 15 adversarial customer scenarios test for vulnerabilities
- Fix it for me: automatically rewrites rule sheet and tests against saved situations
- Train my bot: 30-minute automated training on weak spots with test validation
- GitHub integration: sends improvements as pull requests for engineer approval
- Real chat analysis: identifies problems in actual customer conversations with privacy masking
- Knowledge check: validates bot replies against FAQ and policy documents
- AI comparison: test bot performance across Claude, ChatGPT, and Gemini with costs and speed
- Open source engine: public GitHub implementation of scoring and testing algorithms
|
|---|
| What makes it different | - Multilingual support across 23 European languages in a single 80MB model
- Complete privacy with no data leaving the device
- Free access up to 100k monthly active devices with unlimited inference
| - Nothing ships without engineer approval: all changes go through GitHub pull request review
- Privacy by default: customer data anonymized, keys encrypted, uploads not stored
- Open source testing engine: full algorithm transparency available on GitHub
- Works with your existing AI: supports Claude, ChatGPT, and Gemini with user's own keys
- Unified team experience: one plan covers support, product, and engineering roles
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 38 ▲ +2 (up) since last check | Not checked yet ● no change since last check |
|---|
| Trust Flow (Majestic) | 20 | 0 |
|---|
| Citation Flow (Majestic) | 39 | 0 |
|---|
| Referring domains (Majestic) | 616 | 0 |
|---|
Toxic
Rulebook