| What it does | On-device moderation that flags hateful, abusive and threatening text in 23 European languages, before the message is sent.
| Reliability monitoring for autonomous agent systems
|
|---|
| Stage | MVP
| Not given
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | NL
| Not given
|
|---|
| Pricing model | Freemium
| Paid
|
|---|
| Pricing | Free up to 100k monthly active devices with unlimited inference. Custom licenses available.
| Fixed-fee pilot of about a month credited against the first year if continuing. Annual subscription after pilot in three tiers: Map, Gate, or Continuous.
|
|---|
| 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
| - Detects agents built but never connected
- Identifies work that silently stops between agents
- Catches judges that grade incorrectly
- Prevents changes shipping past broken tests
- Detects identity and data drift
- Measures self-repair behavior
- Works with any model and framework
- Read-only connector with no system changes
- Deployment pipeline gate with pass/fail status
- Encrypted data transmission and on-premises options
|
|---|
| 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
| - Finds failures that never raise an alarm
- Built and proven on real production systems, not theory
- Read-only connector never changes your system
- Works with any model without depending on model vendor
- First findings within a week of pilot start
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | 37 ● no change 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
Obsilient