The NVIDIA Jetson AGX Orin is a state-of-the-art AI computing platform designed to meet the demands of robotics, autonomous machines and edge computing applications.
NVIDIA Jetson AGX Orin Technical Specs
Component | Details |
GPU | NVIDIA Ampere architecture with 2048 CUDA cores and 64 Tensor Cores |
CPU | 12-core Arm Cortex-A78AE v8.2 64-bit CPU |
AI Performance | Up to 275 TOPS |
Memory | 64 GB 256-bit LPDDR5 with 204.8 GB/s bandwidth |
Storage | 64 GB eMMC 5.1 |
Video Encode | 2x 4K60, 4x 4K30, 8x 1080p60, 16x 1080p30 |
Video Decode | 1x 8K30, 2x 4K60, 4x 4K30, 9x 1080p60, 18x 1080p30 |
Power Consumption | Configurable between 15W and 60W |
Form Factor | 100 mm x 87 mm |
I/O Interfaces | 22x PCIe Gen4 lanes, USB 3.2 Gen2, Ethernet, MIPI CSI-2, GPIO |
Operating Temperature | -25°C to 80°C |
Looking to buy NVIDIA Jetson AGX Orin GPUs?
NVIDIA Jetson AGX Orin Key Features
High AI Performance: Up to 275 TOPS for executing complex AI models, including 3D perception and multi-sensor fusion.
Advanced GPU Architecture: Built on NVIDIA Ampere architecture, delivering high performance and efficiency.
Flexible Power Configuration: Configurable between 15W and 60W for energy efficiency or maximum performance.
Comprehensive I/O Support: Includes PCIe Gen4, USB 3.2, and camera inputs for versatile system integration.
Robust Software Ecosystem: Powered by NVIDIA JetPack SDK with tools for AI, computer vision, and deep learning.
NVIDIA Jetson AGX Orin Configurations
Model | AI Performance | GPU Cores | Tensor Cores | Memory | Power Consumption |
Jetson AGX Orin 64GB | Up to 275 TOPS | 2048 | 64 | 64 GB | 15-60W |
Jetson AGX Orin 32GB | Up to 200 TOPS | 1792 | 56 | 32 GB | 15-50W |
Key Differences:
Memory: The 64GB model supports large AI models, while the 32GB model is more cost-effective for moderate workloads.
Performance: The 64GB model offers the highest performance, making it ideal for advanced AI applications.
Power Consumption: Both configurations have configurable power modes but differ in their peak power requirements.
NVIDIA Jetson AGX Orin Use Cases
Industry | Applications |
Robotics | Enables perception and decision-making in autonomous robots. |
Edge AI | Processes real-time data at the edge, reducing latency and bandwidth usage. |
Healthcare | Powers medical imaging and diagnostics. |
Smart Cities | Enhances video analytics for traffic and surveillance systems. |
Autonomous Vehicles | Facilitates object recognition and navigation systems. |
Industrial Automation | Manages robotics for manufacturing and quality control. |
NVIDIA Jetson AGX Orin Additional Capabilities
The NVIDIA Jetson AGX Orin is built for AI applications, featuring 10GbE Ethernet and Time-Sensitive Networking (TSN) for low-latency communication. It includes an NVMe SSD slot for fast, expandable storage and supports tools like DeepStream SDK for video analytics, Triton Inference Server for AI deployment, and Nsight Systems for performance tuning.
It operates in rugged environments from -25°C to 80°C with passive or active cooling. Its Ubuntu-based Linux OS and JetPack SDK enable AI and multimedia development, while USB, PCIe, and Ethernet interfaces ensure easy peripheral integration.
Priced at $1,599 (64GB) and $999 (32GB), it outperforms competitors like Intel Movidius and Google Coral AI. The 64GB model offers 275 TOPS for robotics and advanced AI, while the 32GB version suits moderate workloads like industrial automation. Both support NVIDIA Clara AGX for healthcare and Isaac SDK for robotics.
NVIDIA Jetson AGX Orin Benchmarks
Use Case | Performance Achieved |
Object Detection | 275 FPS (YOLOv5) |
NLP Inference | 2.5 ms per query |
Multi-Camera Inference | 20 streams @ 1080p |
Looking to buy NVIDIA Jetson AGX Orin GPUs?
Comments