Your AI agent for high-stakes vision.

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. Every decision comes with its reasoning. Demonstrated July 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.

Operator training

Train and condition front-line officers.

Defect detection & QC

Maintain consistent quality and process.

Remote review

Analyze rich visual data from anywhere.

Adversarial testing

Exploit and red-team visual AI.

Announcements

News

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).

News

Cignal Teaches Applied AI Auditing and Defense at Black Hat USA 2026

June 2026 - Cignal has been selected to teach Applied AI Auditing and Defense at Black Hat USA 2026.

News

Cignal Selected to Present Research on Vision Language Models (VLMs)

April 06, 2026. - Cignal has been selected to present at SPIE ADIX X. Cignal's research on Vision Language Models is entitled: Let's be reasonable: why compositional VLMs are the future of ATR.