Vertical Data

Contact Us

Vertical Data

Vbox Edge

Overview

The VBox Edge is a high-performance AI server designed for vision-based object detection AI workloads, powered by the Untether AI SpeedAI custom accelerator. Engineered to deliver H100-class performance at a fraction of the cost and power draw, VBox Edge is ideal for AI inference, real-time processing, and edge deployments. Built in collaboration with Dell, this turnkey AI server eliminates the need for multichip partitioning and supports CUDA-based AI models out of the box.

Key Features & Specifications

  • Optimized for AI Vision object detection Workloads – Runs standard CUDA-built AI models (CNNc) efficiently on custom Untether AI SpeedAI accelerators.

  • H100-Class Performance at Lower Cost & Power – Achieves comparable AI inference throughput with significantly improved energy efficiency.

  • Edge-Ready & Flexible Deployment – Available in server, workstation, and custom edge deployments.

  • No Multichip Partitioning Needed – Simplifies AI model execution with a single-chip inference architecture.

  • Multi-Vendor Hardware Support – Financing available for any tier-1 OEM (e.g. Dell, HP, Supermicro).

  • Upcoming AI Software Stack – Orchestration, deployment, and self-serve workload management software arriving soon.

 

Performance & Scalability

  • 5x Improvement in Cost Efficiency – Optimized for inference/s/$ to maximize AI workload ROI.

  • Superior Compute Density – Supports complex semantic segmentation for high-precision defect detection and other vision AI tasks.

  • High Throughput & Energy Efficiency – Processes 3000 images/sec at low power, high throughput.

  • Scalability Without Bottlenecks – Runs inference workloads without bandwidth limitations due to at-memory compute architecture.

Security & Compliance

  • Enterprise-Grade Functional Safety – Supports concurrent, real-time multi-model inference and is ASIL-B certified.

  • Classified AI Model Development – Enables secure inference for defense and government applications without vendor involvement.

  • Reliable Hardware & Software Stack – Built for mission-critical AI deployments, ensuring stability and long-term reliability.

Integration & Compatibility

  • Plug-and-Play AI Model Deployment – Supports FP16 AI models for enhanced inference accuracy

  • Interoperability with ML Frameworks – Seamless integration with TensorFlow, PyTorch, and ONNX using the imAIgine SDK.

  • Cloud & Edge Deployment Options – Deploy in enterprise data centers, at the edge, or in hybrid cloud environments.

  • Optimized Compiler & SDK – The imAIgine SDK enables automated model ingestion, quantization, and push-button deployment.

Industries & Use Cases

  • Vision Systems & Industrial AI – High-speed defect detection, quality assurance, and object recognition.

  • Government & Defense (A&D) – AI-powered situational awareness, ISR (Intelligence, Surveillance, and Reconnaissance), and automated analysis.

  • Autonomous Vehicles – Supports real-time sensor fusion and AI-driven perception for ADAS & self-driving systems.

  • AI-Based Preventative Maintenance – Enhances predictive failure detection and proactive maintenance in industrial settings.

  • Recycling & Smart Sorting – AI-powered real-time object detection for automated waste sorting.

  • Agricultural Technology (AgTech) – AI-driven crop monitoring, pest detection, and farm automation.

  • Software-Defined Vehicles – Advanced AI vision capabilities for next-generation automotive AI applications.

  • Drone & Counter-Drone Systems – AI-driven autonomous navigation and threat detection for UAVs.

Vertical Data