Meet the Nx AI Manager

Nx AI Manager empowers you to seamlessly add intelligence to your video solutions with a universal, scalable pipeline for AI video analytics. With streamlined deployment and precise optimization controls, it makes it easy to deploy and manage models across disparate devices at scale.

Technical Documentation arrow_outward

Key Advantages of Nx AI Manager

Unlock the full potential of AI-powered video with Nx AI Manager. This universal AI model pipeline offers seamless deployment and flexible configurations, empowering you to build custom, adaptable AI-powered video solutions that scale with your application's needs.

Universal Model Deployment

Cross-Hardware AI Flexibility

Instantly deploy and scale AI models seamlessly across a range of industry-leading accelerators or combination of disparate accelerators with a universal model pipeline.

Scalable Solutions

Vision AI For Every Scale

Seamlessly scale AI solutions across diverse environments, from individual devices to enterprise-scale operations, ensuring adaptability as your system grows.

Custom Pipeline Configuration

Fine-Tuned AI Control

Tailor AI pipelines with flexible pre- and post-processing configurations, giving you precise control over data handling to meet the unique requirements of your application.

Model Management

Seamless Model Updates at Scale

Remotely manage and update your AI models across a fleet of devices, ensuring your system remains up-to-date with evolving operational requirements, all from a single interface.

Nx AI Manager Technical Specifications

terminal

Software Requirements

Dual-core Intel i5 quad core 10th gen equivalent X86_64 CPU with at least 4GB of RAM
Raspberry Pi 4b equivalent or higher AARCH64 CPU with at least 4GB of RAM

OS Requirements
Ubuntu Linux LTS version 20.04 or 22.04

Software Requirements
Nx AI Manager 4.0+ [X86_64, Arm32, Arm64]
Nx Meta Client 6.0+ [X86_64, Arm32, Arm64]
Nx Meta Server 6.0+ [X86_64, Arm32, Arm64]

memory

Hardware Requirements

Memory Requirement
It is recommended to add 512 MB in RAM capacity per stream for which the AI plugin is enabled

CPU/GPU Requirement
Up to 6 cameras with AI detection: Intel i5 10th gen or AMD Ryzen 5 3000 Quad-Core.

Up to 12 cameras with AI detection: Intel i7 12th gen or AMD Ryzen 7 3000 Quad-Core.

Other hardware
For other hardware architectures such as ARM, Nvidia Jetson Orin, Hailo, etc, please contact info@networkoptix.com

monitoring

Optimal Performance

For optimal performance of the Nx AI Manager, it is recommended to dedicate the system to the Nx Meta Server application with minimal additional loads. A setup with a single camera and minimal configurations functions well on minimal hardware. However, more complex systems with multiple AI models, numerous cameras, and extensive user connections require significantly enhanced hardware.

AI detection is optimal at a maximum input resolution of 512x512, which aligns with model training specifications. It is advised to use this resolution or slightly higher for secondary camera streams, aiming for at least 6 frames per second to ensure effective AI detection. Running the AI Manager on a primary stream is discouraged due to high CPU load and memory usage, which can impact system performance.