NIN-Redact-v1
Keep personal details out of shared text.
Discuss this model ↗Upcoming model. Capabilities describe our plans; weights and inference are not available.
Keep personal details out of shared text.
Detect and mask PII in notes, transcripts and documents before sharing or passing text to another model. Planned coverage includes Indian identifiers.
Find personal details in notes, documents and transcripts.
Mask detected entities before sharing text with another person or model.
Evaluate Indian identifiers alongside names, email addresses and phone numbers.
See the intended workflow.
Illustrative example, authored by NIN. This is not a live model prediction.
Text · fictional details
Send Maya’s meeting notes to maya@example.com.
Redacted text
Send [PERSON]’s meeting notes to [EMAIL].
Redaction needs human review for sensitive material. Entity coverage and missed detections will need evaluation.
On-device / private-hosted
Built around your workflow.
On-device execution in NIN products and private hosting for organisational workflows are intended deployment options.
Details after evaluation.
Model size, hardware requirements, supported formats, licensing and measured performance will be documented when available. No release date is announced.
Where could NIN-Redact-v1 help?
Tell us about your workflow, the data it uses and what a useful result would look like. Discuss an evaluation partnership with NIN.
Discuss this model ↗