Your agent for high-stakes vision AI.

Describe what you need. Build the scene, or ask what's in it.

4/26 release now available

Your request has been received. Someone will reach out to coordinate your evaluation shortly.

A .gov or .mil address is required.

What you need to see doesn't exist yet.

Work with Cignal to build what you need, then iterate with vision agents to see what works and what doesn't.

Prompt it. See it.

Prompt a multi-modal scene: physically accurate, fully labeled. Threats that don't exist yet. Infrequent defects.

AI that shows its work.

Cignal Ratio: the first compositional VLM in security screening based on our seminal research paper published in 2026 ↗

Work with volumes anywhere

Segment over the network. View on a phone. 100M+ voxels, no install, no workstation. Patent-pending volume-as-video makes both possible.

Try it yourself

Applications

Threat detection

Improve false alarm rates & performance.

Manifest reconciliation

Combat fraud, smuggling, and diversion.

Dynamic assessment

Train and condition front-line officers.

Defect detection & QC

Maintain consistent quality and process.

Operator support

Visual decision support for officers and first responders.

Adversarial testing

Exploit and red-team visual AI.

Announcements

Cignal Launches Tensorfire at Black Hat 2026

August 3, 2026 - Cignal Launches Tensorfire AI Security Testing Distribution at Black Hat 2026.

Cignal Powers Newly Launched DHS S&T Synthetic Baggage Image Dataset to Accelerate Airport Screening Innovation

July 23, 2026 - Cignal Powers Newly Launched DHS S&T Synthetic Baggage Image Dataset to Accelerate Airport Screening Innovation.

Cignal Demonstrates Ratio, the First Compositional VLM for Security Screening

July 2, 2026 - Cignal Demonstrates Ratio, the Security Screening Industry's First Compositional Vision Language Model (VLM).