Industry’s first real-time deep learning platform

Cybersecurity has become an arms race

Hackers have weaponized AI to accelerate the distribution and creation of threats. Malware is proliferating, at a rate of more than 350,000 new malware variants a day.

It’s time to fight back with more intelligent threat defense. How about a deep learning threat brain that not only detects threats it’s been trained on, but also new variants and zero days? More importantly, we do this in less than a second at greater than 99.5% efficacy.

Why deep learning for cybersecurity?

Deep learning is a subset of machine learning that uses multi-layered artificial neural networks to learn data representation. Deep learning algorithms can build more effective models from large amounts of data compared to traditional machine learning algorithms. Our deep learning models have been trained on terabytes of threat data and optimized to deliver inference on 10G traffic.

Watch our lightboard video for more information

Blue Hexagon platform

Combining HexNet™, our proprietary architecture of deep learning neural nets, with our pre-trained AI threat models, our real-time deep learning platform can detect known and unknown network threats in less than a second. Unlike IPS and sandboxes with detection shortcomings and delays, Blue Hexagon delivers efficacy at wire speed without requiring any human triage.

How it works

The Blue Hexagon Platform includes two components:


Deep learning cloud

Daily and diverse threat data is used to train, optimize and validate our deep learning models in our cloud.


Optimized deep learning models are deployed on premises–as virtual machines or hardware appliances to perform deep learning inspection.


We’ve designed our Real-time Deep Learning Platform to deliver speed of detection, efficacy and coverage that sets a new standard for cyberdefense.

Real-time threat inference

Speed matters. Our platform delivers threat detection in less than a second. By integrating with security products you’ve already invested in, we can enable prevention to network devices, firewalls and endpoints. As a result, we’re able to stop the very first victim in the organization from being compromised and stop lateral movement.

Unmatched efficacy for all threats

Security teams can now rely on just one platform to detect both known and known threats. Our platform delivers on average 99.8% detection rates for famous malware families. This includes common attacks like trojans, droppers and ransomware to new ones like fileless attacks. Our platform has also successfully detected new zero day variants such as Anatova and Trickbot.

Gain global threat visibility

The Blue Hexagon Global Threat Cloud incorporates deep learning to classify and curate threats from various intelligence sources. This data is not only used to validate and optimize our models continuously, but also shared with all customers to deliver predictive intelligence into the types of attacks that are targeting specific industries.

Comprehensive kill chain visibility

Most attacks begin with email and moves laterally into the network. When a threat is detected, we deliver complete categorization of the threat, where it fits into the kill chain, along with the indicators of compromise. Our kill chain visualizer enables security operations teams to dive deeper into threats found for additional threat intelligence analysis and threat hunting.

Deployment-ready out of the box

Our solution simply works out of the box. Our deep learning models are optimized and pre-trained in our cloud. This means that as soon as they are deployed on premises, they are ready to get to work to secure your network. Our solution does not require any local tuning or learning.

Enable more from existing security ecoSystem

Your existing security solutions will benefit from our real-time deep learning platform. Because we detect malware in less than a second, we can enable prevention to your existing security products — network devices, endpoints and firewalls.

Ready to get started?