{"title":"Nvidia Products","description":"\u003cp\u003eNVIDIA Jetson is a leading platform of compact, high-performance AI modules designed for edge computing. These modules combine powerful GPU acceleration with energy-efficient architecture, enabling real-time processing of AI workloads directly on devices.\u003c\/p\u003e\n\u003cp\u003eJetson modules are built to power next-generation applications such as robotics, autonomous machines, computer vision, and intelligent automation—delivering server-class AI performance in a small, embedded form factor.\u003c\/p\u003e\n\u003cp\u003eSupported by NVIDIA’s advanced AI software ecosystem, including CUDA, TensorRT, and JetPack, Jetson provides a seamless path from development to deployment.\u003c\/p\u003e","products":[{"product_id":"nvidia-jetson-orin-nano-module","title":"NVIDIA Jetson Orin Nano Module","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;color:#e0e0e0;\"\u003eNVIDIA Jetson Orin Nano Module — Up to 40 TOPS Edge AI — 1024-Core Ampere GPU — CUDA \u0026amp; JetPack 6 Ready\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe \u003cstrong\u003eNVIDIA Jetson Orin Nano Module\u003c\/strong\u003e is a compact AI System-on-Module (SoM) engineered for real-time edge inferencing across robotics, industrial vision, autonomous machines, and AIoT. Available in 4GB (20 TOPS) and 8GB (40 TOPS) configurations, it connects to any Jetson Orin Nano-compatible carrier board via its standard \u003cstrong\u003e260-pin SO-DIMM interface\u003c\/strong\u003e, giving developers, research labs, and OEMs a scalable, power-efficient path from prototype to production deployment.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eUp to 40 TOPS AI Throughput\u003c\/strong\u003e — The 8GB variant delivers 40 TOPS of sparse INT8 inference, enabling real-time object detection, segmentation, and pose estimation entirely at the edge — no cloud round-trip or connectivity dependency required.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNVIDIA Ampere GPU Architecture\u003c\/strong\u003e — 1024 CUDA cores and 32 Tensor Cores on the 8GB (512 cores \/ 16 Tensor Cores on the 4GB) provide hardware-accelerated parallel compute purpose-built for deep learning vision and inference workloads.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHigh-Bandwidth LPDDR5 Memory\u003c\/strong\u003e — 8GB of 128-bit LPDDR5 at 68 GB\/s sustains simultaneous multi-model and multi-stream AI pipelines; the 4GB variant delivers 64-bit LPDDR5 at 34 GB\/s for lighter, single-model workloads.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e6-Core Arm Cortex-A78AE CPU\u003c\/strong\u003e — A high-reliability 64-bit v8.2 CPU running at up to 1.5 GHz handles Linux OS management, sensor preprocessing, and peripheral control in parallel with GPU inference — without performance contention.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMulti-Camera Input Pipeline\u003c\/strong\u003e — Up to 4 simultaneous cameras (expandable to 8 via virtual channels) over 8-lane MIPI CSI-2 with D-PHY 2.1 at up to 20 Gbps aggregate bandwidth — essential for surround-view robotics and multi-sensor fusion systems.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePCIe Gen3 High-Speed Expansion\u003c\/strong\u003e — One x4 plus three x1 PCIe Gen3 lanes allow direct attachment of NVMe SSDs, Wi-Fi 6 adapters, AI accelerator cards, and other high-bandwidth peripherals without USB overhead or bottleneck.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFull CUDA, TensorRT \u0026amp; JetPack Ecosystem\u003c\/strong\u003e — Native support for CUDA, TensorRT, PyTorch, TensorFlow, and NVIDIA JetPack SDK means existing AI models deploy with minimal porting effort and maximum hardware utilisation out of the box.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable Power Envelopes\u003c\/strong\u003e — Selectable 7W, 15W, and 25W power modes let you tune performance vs. thermal budget — critical for battery-powered field devices, passively cooled enclosures, or full-performance embedded applications.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHardware-Accelerated Video Decode\u003c\/strong\u003e — A dedicated NVDEC engine handles up to 1× 4K60 or 11× 1080p30 H.265 streams concurrently, freeing CPU and GPU resources entirely for AI inference tasks running in parallel.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEcosystem-Wide SO-DIMM Compatibility\u003c\/strong\u003e — The 69.6 × 45 mm module with a 260-pin SO-DIMM connector is pin-compatible with a wide range of third-party carrier boards from Connect Tech, Seeed Studio, Auvidea, and others — accelerating custom system integration and time-to-market.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin Nano 4GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin Nano 8GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e20 TOPS\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e40 TOPS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e512-core NVIDIA Ampere, 16 Tensor Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024-core NVIDIA Ampere, 32 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMax GPU Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.0 GHz\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.1 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e6-core Arm Cortex-A78AE v8.2 64-bit, up to 1.5 GHz — 1.5MB L2 + 4MB L3\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e4GB 64-bit LPDDR5 — 34 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR5 — 68 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eExternal NVMe SSD via M.2 M-key (PCIe Gen3) — no internal eMMC\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCamera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eUp to 4 cameras (8 via virtual channels) — 8-lane MIPI CSI-2, D-PHY 2.1 up to 20 Gbps\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1080p30 software-based (CPU\/FFmpeg) — no dedicated NVENC hardware encoder\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× 4K60 (H.265) — 2× 4K30 — 5× 1080p60 — 11× 1080p30 (hardware NVDEC)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1×4 + 3×1 PCIe Gen3 (Root Port \u0026amp; Endpoint)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e3× USB 3.2 Gen2 (10 Gbps) + 3× USB 2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× GbE (10\/100\/1000 Base-T)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× 4K30 multi-mode DP 1.2 (+MST) \/ eDP 1.4 \/ HDMI 1.4\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOther I\/O\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e3× UART, 2× SPI, 2× I2S, 4× I2C, 1× CAN, DMIC \u0026amp; DSPK, PWM, GPIOs\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7W – 10W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7W – 15W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMechanical\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e69.6 × 45 mm — 260-pin SO-DIMM connector\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich Jetson Orin Nano Is Right for You?\u003c\/h3\u003e\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eBoth variants share the same CPU, I\/O set, and connectivity — the decision is purely about memory bandwidth and AI throughput headroom. The 4GB module is the right fit for single-model inference, lightweight NLP tasks, or memory-constrained embedded designs where the lower power ceiling is an asset. The 8GB module is the clear choice whenever you need to run multiple models simultaneously, process 4K multi-stream video analytics, or push sustained throughput above 20 TOPS in production.\u003c\/p\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eFeature\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin Nano 4GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin Nano 8GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eRAM\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e4GB 64-bit LPDDR5\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR5\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory Bandwidth\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e34 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e68 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e20 TOPS\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e40 TOPS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e512 CUDA cores, 16 Tensor Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024 CUDA cores, 32 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower Range\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7W – 10W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7W – 15W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eBest For\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSingle-model inference, power-constrained designs, lightweight NLP \u0026amp; vision tasks\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMulti-model pipelines, 4K multi-stream analytics, demanding real-time inference\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge AI Inference\u003c\/strong\u003e — Run YOLO, ResNet, EfficientDet, and transformer-based models entirely on-device, eliminating cloud latency and data privacy risks in real-time decision systems deployed in the field.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eRobotics \u0026amp; AMR \/ AGV Navigation\u003c\/strong\u003e — Multi-camera input and low-latency Ampere GPU inference make the Orin Nano the ideal compute brain for autonomous mobile robots, warehouse AGVs, and collaborative robot arms requiring real-time environmental perception.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Computer Vision \u0026amp; Defect Inspection\u003c\/strong\u003e — Detect surface defects, measure component tolerances, and classify assembly errors on high-speed production lines with sub-millisecond TensorRT-optimised inference — no cloud connectivity required.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAI-Powered Video Surveillance\u003c\/strong\u003e — Process up to 4 simultaneous camera feeds onboard with person detection, anomaly recognition, and license plate identification — without transmitting raw footage off-site or compromising data privacy.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge NLP \u0026amp; Conversational AI\u003c\/strong\u003e — Deploy compact language models and voice-processing pipelines locally on service kiosks, retail assistants, and point-of-care devices where cloud connectivity is unreliable, costly, or prohibited.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical Imaging \u0026amp; Point-of-Care Diagnostics\u003c\/strong\u003e — Run inference on X-ray, ultrasound, and digital pathology images at the bedside or in remote clinics using a certified hardware-grade module with defined, repeatable power envelopes.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eRetail Analytics \u0026amp; Loss Prevention\u003c\/strong\u003e — Count customers, monitor shelf occupancy, detect unusual behaviours, and generate heatmaps in real time — all processed on-device to comply with GDPR and regional data protection regulations.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrone \u0026amp; UAV Payloads\u003c\/strong\u003e — The 7W minimum power mode and compact SoM footprint allow integration into weight-critical UAV payloads for real-time aerial AI, infrastructure inspection, and autonomous mapping missions.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Agriculture \u0026amp; Precision Farming\u003c\/strong\u003e — Detect plant disease, count crops, and guide autonomous sprayer systems using vision models running on solar or battery-powered field units with no internet dependency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAI Research \u0026amp; Education\u003c\/strong\u003e — Universities, research labs, and maker communities use the Jetson Orin Nano as an accessible yet production-representative platform for developing, benchmarking, and publishing edge AI architectures.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003e1× NVIDIA Jetson Orin Nano Module (SoM) — your selected configuration (4GB or 8GB)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: a carrier board, heatsink, thermal pad, power supply, cables, NVMe SSD, and any cameras or display adapters are not included with the module. These accessories are sold separately. A compatible carrier board and appropriate power supply are required before the module can be used.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat carrier boards are compatible with the Jetson Orin Nano Module?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Orin Nano Module uses a standard \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e260-pin SO-DIMM connector\u003c\/span\u003e and is compatible with any carrier board designed for the Jetson Orin Nano or Jetson Orin NX family. This includes the official \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Orin Nano Developer Kit\u003c\/span\u003e carrier, plus third-party boards from Connect Tech, Seeed Studio, Auvidea, and Forecr, among others. Always verify that the carrier board vendor explicitly lists support for the Orin Nano module, since the older Jetson Nano carrier boards use an incompatible connector and are not interchangeable. Third-party carriers often add features like multiple Ethernet ports, M.2 expansion, or ruggedised I\/O not found on the reference design.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat power supply does the Jetson Orin Nano Module require?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003ePower is delivered through the carrier board, so the supply specification depends on your carrier design. The official NVIDIA Developer Kit carrier uses a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e9V–20V DC barrel jack\u003c\/span\u003e (centre-positive, 5.5mm\/2.5mm), and a 19V \/ 65W adapter is typically recommended to cover all peripherals. The module itself operates across configurable power modes of \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e7W, 15W, and 25W\u003c\/span\u003e — ensure your supply can sustain the peak wattage of your chosen mode plus headroom for attached peripherals such as SSDs, cameras, and USB devices. Third-party carrier boards may specify different input voltage ranges — always consult the carrier datasheet before selecting a power supply.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhich operating systems and JetPack versions are supported?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Orin Nano Module officially supports \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack 5.x and JetPack 6.x\u003c\/span\u003e. The current production release is JetPack 6.2, based on \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 22.04 LTS\u003c\/span\u003e with Linux Kernel 5.15 — it ships with CUDA, TensorRT, cuDNN, OpenCV, VPI, and Multimedia API pre-installed. JetPack 5.x releases are based on Ubuntu 20.04 and remain supported for projects requiring that baseline. Modules shipped with early firmware may require a firmware update before JetPack 6.x can be flashed — NVIDIA's SDK Manager tool handles this process automatically on a Linux host.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available — do I need an SD card or eMMC?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eUnlike the original Jetson Nano, the Jetson Orin Nano Module has \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eno internal eMMC or onboard flash\u003c\/span\u003e — external storage is required to boot. The recommended option is an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD via M.2 M-key (PCIe Gen3)\u003c\/span\u003e, which is the primary boot medium on the official developer kit and delivers the best sustained read\/write performance for AI workloads. Some carrier boards also provide a microSD slot usable for booting or data storage, but NVMe is strongly preferred. Choose an NVMe SSD of at least 32GB; 128GB or larger is recommended to accommodate JetPack, models, datasets, and application logs comfortably.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat additional hardware do I need to get started with this module?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAt minimum you will need: a compatible \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ecarrier board\u003c\/span\u003e, an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD\u003c\/span\u003e (for the OS and JetPack), a suitable \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003epower supply\u003c\/span\u003e matched to the carrier, and a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eheatsink with thermal interface material\u003c\/span\u003e (active cooling is recommended for sustained loads above 15W). For initial setup, a display (HDMI or DisplayPort adapter), USB keyboard, and mouse are useful — though headless setup over SSH is also possible after the first flash. A host PC running Ubuntu 20.04 or 22.04 with NVIDIA SDK Manager installed is required to flash JetPack onto the module for the first time.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson Orin Nano compare to the original Jetson Nano?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Orin Nano is a generational leap beyond the original Jetson Nano (2019). The Orin Nano delivers up to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e40× more AI performance\u003c\/span\u003e, moves from a Maxwell GPU to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Ampere architecture\u003c\/span\u003e, and adds LPDDR5 memory with significantly higher bandwidth. It also gains proper JetPack 5\/6 support on Ubuntu 20.04\/22.04, while the original Jetson Nano reached end-of-life at JetPack 4.6. Critically, the two platforms are physically and software-incompatible — carrier boards, images, and GPIO libraries are not interchangeable — so migrating requires a full system redesign rather than a simple module swap.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO and serial interface pins are available?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe module exposes a rich I\/O set through the 260-pin SO-DIMM connector: \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e3× UART, 2× SPI, 2× I2S, 4× I2C, 1× CAN bus, PWM outputs, and multiple GPIOs\u003c\/span\u003e — plus digital microphone (DMIC) and digital speaker (DSPK) interfaces for audio. The exact signals available to your application depend on your carrier board design; the official developer kit routes a subset to a 40-pin expansion header with a layout broadly familiar to Raspberry Pi users. The CAN bus interface is particularly valuable for robotics and automotive applications requiring deterministic real-time communication with motor controllers and sensors.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs this module suitable for beginners, or is it an advanced platform?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe standalone Jetson Orin Nano Module (SoM) is primarily targeted at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eintermediate-to-advanced users\u003c\/span\u003e — embedded engineers, AI developers, and OEM product teams building custom hardware around the SoM. If you are new to Jetson and want to experiment first, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Orin Nano Developer Kit\u003c\/span\u003e (which bundles a carrier board) is the recommended starting point. Once you are comfortable with JetPack, model deployment workflows, and system integration, the standalone module is the correct choice for designing custom carrier boards or scaling into production. NVIDIA's \"Hello AI World\" and \"Jetson AI Fundamentals\" tutorials, along with the active Jetson Developer Forum community, substantially lower the learning curve.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eDoes the Jetson Orin Nano support hardware video encoding?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThis is one of the most important things to verify before purchasing: the Jetson Orin Nano \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003edoes not include a dedicated hardware video encoder (NVENC)\u003c\/span\u003e — this is a confirmed hardware limitation versus larger Orin variants like the Orin NX and AGX Orin. Video encoding (H.264\/H.265 output streams) must be handled in software via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eCPU cores using FFmpeg or GStreamer\u003c\/span\u003e, limiting simultaneous encode to approximately 1080p30. If your application requires concurrent hardware encoding of multiple high-resolution streams, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Orin NX\u003c\/span\u003e is the appropriate upgrade. Video decode, by contrast, is fully hardware-accelerated via the dedicated NVDEC engine and supports up to 4K60 H.265.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware updates, and community support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAll official documentation, hardware design files, datasheets, and JetPack firmware are hosted on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Developer Zone\u003c\/span\u003e at developer.nvidia.com\/embedded. NVIDIA's \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eSDK Manager\u003c\/span\u003e tool handles JetPack flashing and software updates from a Ubuntu host. Community support, model optimisation discussions, and hardware bring-up guidance are available on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums — Jetson \u0026amp; Embedded Systems\u003c\/span\u003e section, which is actively monitored by NVIDIA engineers. For AI deployment tutorials, the freely available \"Hello AI World\" and \"Jetson AI Fundamentals\" courses are specifically designed around Jetson Orin Nano hardware and are the recommended starting point for new users.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"4GB","offer_id":43114356310121,"sku":"NVD-002","price":28599.99,"currency_code":"INR","in_stock":true},{"title":"8GB","offer_id":43194982760553,"sku":null,"price":31899.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/IMG-0404.jpg?v=1778242970"},{"product_id":"nvidia-jetson-orin-nano-super-developer-kit","title":"NVIDIA Jetson Orin Nano Super Developer Kit","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;color:#ffffff;\"\u003eNVIDIA Jetson Orin Nano Super Developer Kit — 67 TOPS AI Performance — 1024-Core Ampere GPU — 102 GB\/s LPDDR5\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe Jetson Orin Nano Super Developer Kit is the most capable compact edge AI computer NVIDIA has released at this power envelope — delivering \u003cstrong\u003e67 TOPS of INT8 AI performance\u003c\/strong\u003e inside a 7W–25W footprint that fits robots, drones, and field-deployed systems alike. Backed by the full JetPack 6.2 SDK with CUDA 12.6, TensorRT 10.3, cuDNN 9.3, and DLA 3.1, it provides every tool needed to develop and ship production AI models without the cloud.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e67 TOPS On-Device AI — No Cloud Required\u003c\/strong\u003e — Run complex neural networks — object detection, segmentation, LLMs, and NLP — entirely on-device at 67 INT8 TOPS sparse, eliminating cloud latency and recurring inference costs in a single board.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e1024-Core Ampere GPU @ 1020 MHz\u003c\/strong\u003e — 32 Tensor Cores handle mixed-precision inference and training loops at a GPU clock 60% higher than the original Orin Nano, delivering smoother real-time pipelines without stepping up to a larger board.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e102 GB\/s Memory Bandwidth — 50% Wider Than Before\u003c\/strong\u003e — The shared 8GB LPDDR5 pool feeds the CPU, GPU, and DLA simultaneously without bottlenecking multi-model pipelines, enabling richer sensor-fusion architectures on a single compact board.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFlexible 7W – 25W Super Power Mode\u003c\/strong\u003e — The new 25W Super mode unlocks peak throughput for demanding workloads, while 7W and 15W modes extend runtime in battery-powered or thermally constrained field deployments — all switchable at runtime with no hardware changes.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFull JetPack 6.2 SDK — Pre-Integrated \u0026amp; Tested\u003c\/strong\u003e — Ships with CUDA 12.6, TensorRT 10.3, cuDNN 9.3, VPI 3.2, and DLA 3.1 on Ubuntu 22.04 — every component in the NVIDIA AI stack verified together so first-day productivity requires zero manual dependency resolution.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDual 40-Pin GPIO Headers for Sensor Fusion\u003c\/strong\u003e — Two industry-standard expansion headers expose UART, SPI, I2C, I2S, and PWM at 3.3V logic levels, enabling multi-sensor fusion, actuator control, and custom peripheral integration without a separate microcontroller.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSix USB 3.2 Gen 2 Type-A Ports at 10 Gbps Each\u003c\/strong\u003e — Connect depth cameras, USB hubs, storage drives, and lab instruments simultaneously without bandwidth contention — a port density rarely found on embedded AI boards at this form factor and power class.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNVMe SSD Expansion via M.2 Key M\u003c\/strong\u003e — The M.2 Key M socket accepts NVMe SSDs up to 2280 size over PCIe 3.0 ×4, delivering several times the sustained I\/O throughput of microSD for dataset-heavy training, logging, and high-throughput inference pipelines.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNative ROS 2 Humble \u0026amp; Isaac ROS Support\u003c\/strong\u003e — Hardware-accelerated computer vision via VPI and the Isaac ROS Gem library plug directly into the ROS 2 Humble graph, making this the preferred development target for the edge robotics ecosystem.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eDetails\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e67 TOPS (INT8 Sparse) \/ 33 TOPS (INT8 Dense)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024-core NVIDIA Ampere with 32 Tensor Cores @ 1020 MHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e6-core Arm Cortex-A78AE v8.2 64-bit @ 1.7 GHz, 1.5MB L2 + 4MB L3\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR5 @ 102 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003emicroSD slot + M.2 Key M (NVMe SSD, PCIe 3.0 ×4, up to 2280)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e6× USB 3.2 Gen2 Type-A (10 Gbps) + 1× USB 3.2 Type-C (Host \/ Device \/ Recovery)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× DisplayPort 1.2\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGigabit Ethernet\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eWireless\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eM.2 Key E Wi-Fi \/ Bluetooth (pre-installed)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCamera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× MIPI CSI-2 connectors (up to 4-lane)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPIO \u0026amp; Expansion\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× 40-pin headers (UART, SPI, I2C, I2S, PWM)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7W – 25W (configurable power modes)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSoftware Stack\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eJetPack 6.2+, Ubuntu 22.04, CUDA 12.6, TensorRT 10.3, cuDNN 9.3, VPI 3.2, DLA 3.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Mobile Robots\u003c\/strong\u003e — Fuse lidar, depth camera, and IMU data while running concurrent SLAM and path-planning models on a single board compact enough to fit inside a tabletop mobile chassis.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Camera \u0026amp; Vision Systems\u003c\/strong\u003e — Deploy multi-stream object detection, tracking, and licence-plate recognition at the camera edge, keeping raw video on-premises and eliminating bandwidth costs to the cloud entirely.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrone \u0026amp; UAV Payload Processing\u003c\/strong\u003e — Perform real-time aerial image segmentation and obstacle-avoidance inference onboard without ground-station uplinks or cellular data, enabling fully autonomous flight in GPS-denied environments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Automation \u0026amp; Quality Inspection\u003c\/strong\u003e — Run visual defect-detection models directly on the factory floor and trigger actuators or alerts via GPIO with sub-millisecond latency, without routing sensitive video to a centralised server.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge NLP \u0026amp; Voice Interfaces\u003c\/strong\u003e — Process wake-word detection, speech-to-text, and intent classification locally — ideal for industrial HMIs, kiosks, and assistive devices with strict data-privacy or offline-operation requirements.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical \u0026amp; Biomedical Imaging\u003c\/strong\u003e — Accelerate portable diagnostic devices that analyse X-ray, ultrasound, or endoscopic imagery with TensorRT-optimised CNNs while keeping patient data completely off external networks.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Agriculture \u0026amp; Field Robotics\u003c\/strong\u003e — Mount on autonomous tractors or sprayers to run crop-disease detection and yield-estimation models in environments with no reliable cellular coverage or cloud uplink.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eRetail \u0026amp; Logistics Analytics\u003c\/strong\u003e — Analyse foot traffic, shelf occupancy, and customer behaviour in real time at the store edge, processing sensitive video locally and delivering insights without raw data leaving the premises.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEducational AI \u0026amp; Robotics Research\u003c\/strong\u003e — A full CUDA + TensorRT environment on a low-power board makes it ideal for university labs, capstone projects, and competitions such as FIRST Robotics and Robomaster.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eROS 2 \u0026amp; Isaac ROS Development\u003c\/strong\u003e — Native ROS 2 Humble support, hardware-accelerated VPI, and the Isaac ROS Gem library make this the go-to development target for building production robotics pipelines on the ROS 2 edge ecosystem.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003eJetson Orin Nano 8GB Super module (pre-installed on reference carrier board)\u003c\/li\u003e\n  \u003cli\u003eActive heatsink with cooling fan (pre-attached)\u003c\/li\u003e\n  \u003cli\u003e19V \/ 45W power supply\u003c\/li\u003e\n  \u003cli\u003ePower cable — Type B (US \/ JP)\u003c\/li\u003e\n  \u003cli\u003ePower cable — Type I (CN)\u003c\/li\u003e\n  \u003cli\u003eQuick Start \u0026amp; Support Guide\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: accessories such as microSD cards, NVMe SSDs, USB keyboards, displays, cameras, and cables are sold separately and not included unless stated above.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat operating systems and software does the Jetson Orin Nano Super Developer Kit support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe kit runs \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack 6.2+\u003c\/span\u003e, built on \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 22.04 LTS\u003c\/span\u003e with Linux kernel 5.15. The software stack includes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eCUDA 12.6\u003c\/span\u003e, \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eTensorRT 10.3\u003c\/span\u003e, cuDNN 9.3, VPI 3.2, DLA 3.1, and the DeepStream SDK — everything needed for end-to-end AI pipeline development. Popular frameworks including PyTorch, TensorFlow, and ONNX Runtime are available as optimised containers from the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA NGC registry\u003c\/span\u003e. ROS 2 Humble is fully supported alongside the Isaac ROS hardware-accelerated Gem library. The platform also supports Triton Inference Server for multi-model serving at the edge.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat power supply does the Jetson Orin Nano Super Developer Kit require?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe kit requires a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e19V DC barrel-jack input\u003c\/span\u003e through the included supply, rated at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e45W\u003c\/span\u003e — enough headroom for the full 25W Super power mode plus USB peripheral draw. Do not substitute a generic USB-C PD adapter or similarly rated laptop charger; voltage regulation on Jetson boards is strict and an incorrect supply will cause random reboots or prevent boot entirely. The barrel connector is a standard 5.5mm \/ 2.5mm centre-positive type — any replacement supply must match this specification exactly. The included supply accepts \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e100–240V AC input\u003c\/span\u003e, making it globally compatible, though a local plug adapter may be needed outside the US, JP, or CN regions. Always power the board from the barrel jack rather than the USB Type-C port for stable operation under full load.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eCan I upgrade the firmware, and how does JetPack updating work?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003e\u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack\u003c\/span\u003e is updated via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e on an Ubuntu host PC, or by reflashing the microSD card or NVMe drive with a fresh image from the Jetson Software Downloads page. The \"Super\" performance boost is a software-level unlock delivered with JetPack 6.2 — meaning owners of the original Jetson Orin Nano 8GB Developer Kit can reach Super performance figures through a firmware upgrade alone, with no new hardware required. Always back up your work before reflashing, as the process overwrites the root filesystem entirely. Incremental apt-based updates can apply minor security patches between major JetPack releases without a full reflash. NVIDIA typically maintains two active JetPack branches at a time, with release notes published on the JetPack SDK page at developer.nvidia.com.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available, and which should I use?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe carrier board provides a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003emicroSD card slot\u003c\/span\u003e and an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eM.2 Key M socket\u003c\/span\u003e for NVMe SSDs in 2230 or 2280 form factor over PCIe 3.0 ×4. For early prototyping, a UHS-I Class 10 \/ A2-rated microSD card at 64GB or larger is sufficient to boot JetPack and run light workloads. For production, dataset-heavy, or logging-intensive applications, an NVMe SSD is strongly recommended — it delivers several times the sequential throughput and far superior sustained I\/O performance, as microSD throttles significantly under prolonged write loads. USB storage is supported but is not suitable as the primary root device due to latency and bus-sharing limitations. For best overall performance, boot from NVMe and reserve the microSD slot as a secondary or recovery medium.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat accessories do I need to get started?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eBeyond what's in the box, you'll need a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003emicroSD card\u003c\/span\u003e (64GB+ recommended) or an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD\u003c\/span\u003e for the OS, a USB keyboard and mouse, a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eDisplayPort monitor\u003c\/span\u003e or a DisplayPort-to-HDMI active adapter, and a USB-A to USB-C cable for initial flashing via recovery mode. A host PC running Ubuntu 20.04 or 22.04 is required for SDK Manager-based firmware flashing — though pre-built microSD images are available for a simpler first-boot experience without the host PC requirement. For camera-based projects, MIPI CSI-2 sensor modules compatible with the Jetson Orin Nano (such as the IMX219 or IMX477 class) connect directly to the two on-board CSI connectors. An NVMe SSD is optional but strongly recommended for any dataset storage or AI training workload. All accessories are sold separately and are not included with the kit.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Orin Nano Super compare to the original Jetson Orin Nano?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Super achieves \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e67 TOPS\u003c\/span\u003e versus the original's 40 TOPS — a 67% improvement — driven by a GPU clock increase from 635 MHz to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e1020 MHz\u003c\/span\u003e and a memory bandwidth uplift from 68 GB\/s to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e102 GB\/s\u003c\/span\u003e. A new \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e25W Super power mode\u003c\/span\u003e provides additional thermal headroom not available on the original, which topped out at 15W. The physical module and carrier board are pin-compatible, and the Super performance figures are available to original 8GB Orin Nano hardware owners through a JetPack 6.2 firmware upgrade — no hardware swap required. The improvement is particularly impactful for generative AI workloads such as LLMs, VLMs, and Vision Transformers, which are memory-bandwidth-bound and benefit nearly linearly from the 50% bandwidth gain. There are no changes to the CPU core count, I\/O complement, or board form factor.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO pins are available, and what interfaces do they expose?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe developer kit provides \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003etwo 40-pin GPIO expansion headers\u003c\/span\u003e exposing UART, SPI, I2C, I2S, PWM, and general-purpose digital I\/O at 3.3V logic levels. The pinout is fully documented in the official Jetson Orin Nano Developer Kit User Guide and is partially compatible with Raspberry Pi HATs. Additionally, the board provides \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e2× MIPI CSI-2 camera connectors\u003c\/span\u003e (up to 4-lane each), six USB 3.2 Gen 2 Type-A ports, one USB Type-C, Gigabit Ethernet, DisplayPort 1.2, and a pre-installed M.2 Key E Wi-Fi\/Bluetooth module. The MIPI CSI-2 connectors support sensors up to the IMX477 12MP class, making them suitable for stereo vision, depth, and hyperspectral imaging rigs. GPIO control in Python is handled via the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson.GPIO\u003c\/span\u003e library, which follows a convention similar to RPi.GPIO for straightforward migration from Raspberry Pi projects.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs this kit suitable for beginners, or is it aimed at advanced developers?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Orin Nano Super Developer Kit caters to both audiences effectively. Beginners can follow NVIDIA's \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eHello AI World\u003c\/span\u003e tutorial series to run pre-trained object-detection and image-classification models within minutes using Python and the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ejetson-inference\u003c\/span\u003e library — no prior C++ or embedded Linux experience required. Advanced developers have full access to CUDA kernel development, TensorRT custom plugin authoring, the DeepStream SDK for multi-stream video analytics, and direct hardware register access for custom MIPI and GPIO peripherals. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNGC container registry\u003c\/span\u003e provides optimised baseline containers for PyTorch, TensorFlow, and Triton that eliminate the most time-consuming environment setup steps. The active NVIDIA Developer Forums and the JetsonHacks community offer fast resolutions to common issues at any experience level.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat's a common mistake to avoid when setting up the Jetson Orin Nano Super?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most frequent setup mistake is using an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eunderpowered or non-compliant power supply\u003c\/span\u003e — the board requires a stable 19V source, and generic USB-C PD adapters or unregulated laptop chargers will cause random reboots or fail to boot under load. A second common issue is running MAXN (25W Super) mode \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ewithout adequate airflow\u003c\/span\u003e — the pre-attached fan must be connected and unobstructed to prevent thermal throttling under sustained AI workloads. A third pitfall is flashing an older JetPack 5.x image instead of JetPack 6.x, which does not include the Super power-mode unlock and will silently cap performance at the original 40-TOPS profile. Always verify the active power mode with \u003ccode\u003esudo nvpmodel -q\u003c\/code\u003e and monitor thermals live with \u003ccode\u003etegrastats\u003c\/code\u003e. Confirm the power mode is correct after every reflash, as it does not persist across full image reinstalls.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware updates, and community support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eOfficial documentation lives at the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Developer Portal\u003c\/span\u003e (developer.nvidia.com\/embedded), including the Orin Nano Developer Kit User Guide, full pinout diagrams, and JetPack release notes. Firmware and SDK downloads are managed through \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e or the Jetson Software Downloads page at developer.nvidia.com\/embedded\/downloads. For community support, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums\u003c\/span\u003e (Jetson \u0026amp; Embedded Systems category) are the most active resource, with NVIDIA engineers regularly participating in technical threads. The JetsonHacks blog and associated GitHub repositories provide community-maintained tutorials for hardware add-ons, custom carrier boards, and popular AI framework integrations. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA NGC catalogue\u003c\/span\u003e at ngc.nvidia.com hosts official pre-built containers for PyTorch, TensorFlow, Triton, and DeepStream that are tested and validated against each JetPack release.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":42896621305961,"sku":"NVD-004","price":30999.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/H76260cead4e14aaf9339c34b77d18b75k.avif?v=1770658279"},{"product_id":"nvidia-jetson-orin-nx-module","title":"NVIDIA Jetson Orin NX Module","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;color:#e0e0e0;\"\u003eNVIDIA Jetson Orin NX 8GB — 70 TOPS Edge AI — 1024-Core Ampere GPU — Compact SO-DIMM Module\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe \u003cstrong\u003eNVIDIA Jetson Orin NX 8GB\u003c\/strong\u003e delivers up to \u003cstrong\u003e70 TOPS\u003c\/strong\u003e of INT8 AI performance — scaling to 117 TOPS in MAXN_SUPER mode with JetPack 6.2 — within the ultra-compact 69.6×45mm SO-DIMM form factor. It combines a 1024-core NVIDIA Ampere GPU, 6-core Arm Cortex-A78AE CPU, dedicated NVDLA v2.0 deep learning acceleration, and hardware video encode\/decode into a configurable 10W–40W thermal envelope, making it the go-to compute module for drones, autonomous robots, smart cameras, and portable industrial AI systems. Available alongside the Orin NX 16GB for deployments demanding greater memory capacity and additional CPU cores.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e70 TOPS AI Performance — 117 TOPS in MAXN_SUPER Mode\u003c\/strong\u003e — delivers real-time deep learning inference for vision, NLP, and multi-sensor fusion pipelines; JetPack 6.2 unlocks MAXN_SUPER for a 67% throughput boost without a hardware change.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e1024-Core NVIDIA Ampere GPU with 32 Tensor Cores\u003c\/strong\u003e — mixed-precision matrix operations (FP32, FP16, INT8, INT4) accelerate neural network layers at sub-millisecond latency, enabling concurrent multi-model inference on a single module.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e1× NVDLA v2.0 Deep Learning Accelerator\u003c\/strong\u003e — offloads steady-state inference from the GPU at up to 20 TOPS, freeing Ampere cores for pre\/post-processing, sensor fusion, and secondary model workloads running in parallel.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e6-Core Arm Cortex-A78AE CPU at 2 GHz\u003c\/strong\u003e — automotive-grade 64-bit cores with hardware ECC handle ROS2 node execution, sensor preprocessing, and Linux OS management without thermal throttling under sustained workloads.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e8GB 128-Bit LPDDR5 at 102.4 GB\/s\u003c\/strong\u003e — wide memory bandwidth prevents bottlenecks when streaming multiple high-resolution camera feeds or running concurrent multi-model inference — double the bandwidth of the previous Xavier NX generation.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHardware Video Encode \u0026amp; Decode\u003c\/strong\u003e — a dedicated VPU handles 4K60 H.265 encoding and 8K30 H.265 decoding entirely in hardware, preserving CPU and GPU resources for AI inference tasks running simultaneously.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eUp to 4 MIPI CSI-2 Cameras (8 via Virtual Channels)\u003c\/strong\u003e — 8 D-PHY 1.2 lanes at 20 Gbps aggregate support stereo depth rigs, 360° vision arrays, and simultaneous RGB\/thermal capture without an external frame grabber.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePCIe Gen 4 Connectivity (1×4 + 3×1 lanes)\u003c\/strong\u003e — supports high-throughput NVMe SSDs, FPGA accelerators, and multi-port networking cards directly on the module bus, with substantially lower latency than USB-attached peripherals.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable TDP: 10W to 40W\u003c\/strong\u003e — four selectable power modes let you tune performance against battery life or thermal budget at deployment time, with no firmware reflash required when switching between modes.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eXavier NX Carrier Board Compatible\u003c\/strong\u003e — the 260-pin SO-DIMM interface is pin-compatible with Jetson Xavier NX carrier boards, minimising redesign effort and bill-of-materials changes when migrating existing platforms to Orin-class performance.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eDetails\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e70 TOPS (INT8) — 117 TOPS (MAXN_SUPER, JetPack 6.2+)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNVIDIA Ampere Architecture — 1024 CUDA Cores, 32 Tensor Cores, 1173 MHz max\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e6-Core Arm® Cortex®-A78AE v8.2 64-bit — 2 GHz max — 1.5MB L2 + 4MB L3\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR5 — 102.4 GB\/s bandwidth\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNo on-module storage — external NVMe SSD via carrier board M.2 slot\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× NVDLA v2.0 (up to 20 TOPS)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× PVA v2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×4K60 | 3×4K30 | 6×1080p60 | 12×1080p30 (H.265) — H.264, AV1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×8K30 | 2×4K60 | 4×4K30 | 9×1080p60 | 18×1080p30 (H.265) — H.264, VP9, AV1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCSI Camera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUp to 4 cameras (8 via virtual channels) — 8 MIPI CSI-2 lanes — D-PHY 1.2 (up to 20 Gbps)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× 8K30 multi-mode DP 1.4a (+MST) \/ eDP 1.4a \/ HDMI 2.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×4 + 3×1 (PCIe Gen 4, Root Port \u0026amp; Endpoint)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e3× USB 3.2 Gen2 (10 Gbps) — 3× USB 2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× Gigabit Ethernet\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOther I\/O\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e3× UART, 2× SPI, 2× I2S, 4× I2C, 1× CAN, DMIC \u0026amp; DSPK, PWM, GPIOs\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10W \/ 15W \/ 25W \/ 40W (MAXN_SUPER, JetPack 6.2+)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMechanical\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e69.6mm × 45mm — 260-pin SO-DIMM connector\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich Jetson Orin NX Is Right for You?\u003c\/h3\u003e\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eBoth the 8GB and 16GB share the same Ampere GPU, PCIe Gen 4 connectivity, and SO-DIMM form factor — the key differentiators are memory capacity, CPU core count, and peak AI throughput. Choose the 8GB for compact single-model deployments and power-sensitive platforms; choose the 16GB when running concurrent models, higher-resolution pipelines, or multi-agent robotic stacks that demand larger working memory.\u003c\/p\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eFeature\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin NX 8GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eOrin NX 16GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e70 TOPS (117 TOPS MAXN_SUPER)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e100 TOPS (157 TOPS MAXN_SUPER)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024 CUDA Cores, 32 Tensor Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024 CUDA Cores, 32 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e6-Core A78AE, 2 GHz\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8-Core A78AE, 2 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB LPDDR5 — 102.4 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e16GB LPDDR5 — 102.4 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× NVDLA v2.0\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× NVDLA v2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower Modes\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10W \/ 15W \/ 25W \/ 40W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10W \/ 15W \/ 25W \/ 40W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eBest For\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDrones, handheld systems, single-model inference, battery-powered platforms\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMulti-model pipelines, robotic arms, high-res video AI, concurrent sensor fusion stacks\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Drones \u0026amp; UAVs\u003c\/strong\u003e — the 10W–25W power envelope and compact SO-DIMM form factor fit directly into UAV flight computers, enabling onboard obstacle avoidance, target tracking, and real-time path planning without a tethered compute unit.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Robot Vision\u003c\/strong\u003e — runs real-time object detection, grasp pose estimation, and defect classification simultaneously, with hardware CSI-2 camera support for synchronised multi-camera bin-picking and assembly verification rigs.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutomated Optical Inspection (AOI)\u003c\/strong\u003e — high-bandwidth LPDDR5 and hardware video decode handle line-scan and area-scan camera feeds at production-line speeds for PCB, semiconductor, and pharmaceutical surface inspection.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eCollaborative Robots (Cobots)\u003c\/strong\u003e — runs ROS2 navigation stacks, sensor fusion middleware, and safety watchdog processes concurrently on dedicated CPU cores while the GPU handles visual odometry and real-time scene understanding.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Traffic \u0026amp; Surveillance Systems\u003c\/strong\u003e — hardware H.265 encode\/decode and DeepStream SDK support multi-stream vehicle tracking, licence plate recognition, and crowd analytics entirely at the edge without cloud connectivity or data egress costs.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHandheld Medical Imaging Devices\u003c\/strong\u003e — configurable TDP and compact form factor enable battery-powered ultrasound, retinal scanning, and dermatology AI tools that require real-time inference without transmitting patient data to the cloud.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge Inference Servers\u003c\/strong\u003e — TensorRT-optimised INT8 models with NVDLA offloading make the Orin NX 8GB a capable low-power inference server for smart factory edge nodes, retail AI kiosks, and digital signage analytics platforms.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNatural Language Processing at the Edge\u003c\/strong\u003e — 8GB LPDDR5 is sufficient to run quantised large language models and speech recognition pipelines for voice-controlled industrial HMIs and autonomous service robot interfaces.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Ground Vehicles (AGVs)\u003c\/strong\u003e — paired with a carrier board providing CAN bus, UART, and GPIO breakouts, the Orin NX 8GB integrates directly into AGV motor controllers, LiDAR processing pipelines, and fleet management stacks.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eResearch \u0026amp; Academic Prototyping\u003c\/strong\u003e — the SO-DIMM carrier board ecosystem, JetPack SDK, and NVIDIA NGC model catalogue provide a well-documented, production-representative platform for robotics labs and universities developing next-generation embodied AI systems.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003e1× NVIDIA Jetson Orin NX 8GB Module (with integrated Thermal Transfer Plate)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: a carrier board, heatsink, thermal pad, power supply, NVMe SSD, cables, cameras, and display adapters are sold separately and not included with the module. A compatible carrier board and appropriate power supply are required before the module can be used.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat carrier boards are compatible with the Jetson Orin NX 8GB?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Orin NX 8GB uses a standard \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e260-pin SO-DIMM connector\u003c\/span\u003e that is pin-compatible with carrier boards designed for the Jetson Xavier NX family, allowing most existing carrier board designs to be reused with a JetPack firmware update. Third-party carriers from Connect Tech, Seeed Studio, Auvidea, and Forecr also explicitly support the Orin NX module. Always verify the carrier board vendor lists Orin NX compatibility, since the older Jetson Nano carrier boards use an incompatible connector and are not interchangeable. A carrier board with an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eM.2 Key M slot\u003c\/span\u003e is strongly recommended, as it is required for NVMe SSD storage which serves as the primary boot and OS medium.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat power supply does the Jetson Orin NX 8GB require?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003ePower is delivered entirely through the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e260-pin SO-DIMM carrier board connector\u003c\/span\u003e — there is no separate power input on the module. The carrier board regulates the supply; most commercial designs accept \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e9V–20V DC input\u003c\/span\u003e, with a 19V \/ 65W adapter commonly used to cover all peripherals. Total system draw in standard modes ranges from 10W to 25W; enabling \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eMAXN_SUPER mode\u003c\/span\u003e with JetPack 6.2 can push peak draw to 40W. For battery-powered designs, budget at least 20W headroom above the chosen TDP to handle transient GPU and CPU burst peaks.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhich operating systems and software frameworks are supported?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Orin NX 8GB runs Ubuntu-based Linux via the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack SDK\u003c\/span\u003e, which bundles CUDA, cuDNN, TensorRT, DeepStream, and VPI in a tested, unified software stack. Both JetPack 5.x (Ubuntu 20.04) and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack 6.x (Ubuntu 22.04 LTS)\u003c\/span\u003e are supported, with JetPack 6.2 adding MAXN_SUPER performance mode. \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ePyTorch, TensorFlow, and ONNX Runtime\u003c\/span\u003e are available via JetPack-aligned wheel packages, and ROS2 Humble and Jazzy integrate natively. Docker containers are supported for isolated pipeline deployments, and NVIDIA's NGC catalogue provides hundreds of pre-optimised models ready for TensorRT deployment.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eDoes the Jetson Orin NX 8GB have built-in storage, and what options are available?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Orin NX 8GB has \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eno on-module flash or eMMC\u003c\/span\u003e — all OS and application storage is provided externally through the carrier board. The recommended option is an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD via M.2 Key M (PCIe Gen 4)\u003c\/span\u003e, which delivers the best sustained read\/write performance for AI workloads, logging-heavy pipelines, and dataset storage. Some carrier boards expose a microSD slot as an alternative, but SD card I\/O will bottleneck demanding workloads. A capacity of at least 64GB is recommended to accommodate JetPack, AI models, application code, and log files comfortably.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat accessories are required to get started with the Jetson Orin NX 8GB?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAt minimum you need a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ecompatible carrier board\u003c\/span\u003e, an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD\u003c\/span\u003e for storage, and a DC power supply matched to the carrier's input specification. A display cable (HDMI 2.1 or DisplayPort 1.4a), USB keyboard, and mouse are recommended for initial setup, along with a host PC running \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e on Ubuntu to flash JetPack. A \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ethermal solution\u003c\/span\u003e — heatsink with thermal interface material and, ideally, an active cooling fan for sustained loads above 15W — is essential; the module includes a Thermal Transfer Plate but requires external cooling hardware to remain within thermal limits.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson Orin NX 8GB compare to the Jetson Xavier NX?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Orin NX 8GB delivers approximately \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e5× the AI throughput\u003c\/span\u003e of the Xavier NX 8GB — from roughly 21 TOPS to 70 TOPS (117 TOPS MAXN_SUPER) — while maintaining the same SO-DIMM form factor for drop-in carrier board reuse. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eAmpere GPU\u003c\/span\u003e replaces the older Volta architecture, adding INT4 precision and significantly improved Tensor Core efficiency per watt. Memory bandwidth doubles from 51.2 GB\/s to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e102.4 GB\/s\u003c\/span\u003e thanks to LPDDR5, and PCIe upgrades from Gen 3 to Gen 4, delivering higher peripheral throughput. Power modes are also more granular on the Orin NX, offering 10W, 15W, 25W, and 40W profiles versus the Xavier NX's 10W and 15W options.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat GPIO and serial interfaces are available on the Jetson Orin NX 8GB?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThrough a compatible carrier board, the Orin NX 8GB exposes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e3× UART, 2× SPI, 2× I2S audio, 4× I2C, 1× CAN bus\u003c\/span\u003e, DMIC and DSPK digital audio, PWM outputs, and multiple configurable GPIO lines. PCIe provides \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e1×4 + 3×1 Gen 4 lanes\u003c\/span\u003e, while USB includes 3× USB 3.2 Gen2 (10 Gbps) and 3× USB 2.0 host ports. The CAN bus interface is particularly valuable for robotics and AGV applications requiring deterministic real-time communication with motor controllers and sensors. The exact signals available depend on your carrier board design — consult the carrier schematic to confirm interface routing before connecting external hardware.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs the Jetson Orin NX 8GB suitable for beginners or is it an advanced platform?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe standalone Orin NX 8GB SoM is primarily aimed at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eintermediate-to-advanced developers\u003c\/span\u003e — embedded engineers, AI developers, and OEM product teams building custom hardware. Initial setup requires flashing JetPack via SDK Manager, configuring carrier board device trees, and managing Linux networking and storage, all of which assume embedded Linux familiarity. Developers new to Jetson should start with the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Orin Nano Developer Kit\u003c\/span\u003e, which ships as a complete kit. Once your environment is established, the Orin NX integrates smoothly with high-level frameworks like \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ePyTorch, DeepStream, and ROS2\u003c\/span\u003e, significantly lowering the barrier for AI model deployment and iteration.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat common mistakes should I avoid when deploying the Jetson Orin NX 8GB?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most common mistake is running at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eMAXN or MAXN_SUPER power mode\u003c\/span\u003e without a properly rated thermal solution — the Orin SoC aggressively throttles CPU and GPU clocks when the die exceeds thermal limits, causing unpredictable latency spikes in production that are difficult to diagnose. Always validate thermals under sustained full-load conditions before finalising an enclosure design. A second frequent issue is \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eattempting to flash without a recognised storage device\u003c\/span\u003e — the module requires a formatted NVMe SSD or compatible storage present on the carrier before SDK Manager can deploy JetPack. Finally, verify your carrier board's \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack compatibility list\u003c\/span\u003e before updating firmware, as some board support packages lag behind the latest JetPack release by several weeks.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware, and community support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eOfficial documentation — including the Orin NX Series datasheet, hardware design guide, and JetPack release notes — is hosted on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Developer Zone\u003c\/span\u003e at developer.nvidia.com\/embedded. Firmware and the NVIDIA SDK Manager installer are available at developer.nvidia.com\/nvidia-sdk-manager. Community support, carrier board integration guides, and TensorRT optimisation discussions are active on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums — Jetson \u0026amp; Embedded Systems\u003c\/span\u003e section, monitored by NVIDIA engineers. Pre-optimised AI models ready for Jetson deployment — covering object detection, segmentation, pose estimation, and NLP — are available through the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA NGC catalogue\u003c\/span\u003e at ngc.nvidia.com.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"8GB","offer_id":43036766208105,"sku":"NVD-007","price":59599.99,"currency_code":"INR","in_stock":true},{"title":"16GB","offer_id":43036766240873,"sku":"NVD-008","price":90049.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/2-2.jpg?v=1774261891"},{"product_id":"nvidia-jetson-xavier-nx-module","title":"NVIDIA Jetson Xavier NX Module","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;color:#e0e0e0;\"\u003eNVIDIA Jetson Xavier NX — 21 TOPS Edge AI — 384-Core Volta GPU — 8GB \u0026amp; 16GB LPDDR4x\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe \u003cstrong\u003eNVIDIA Jetson Xavier NX Module\u003c\/strong\u003e is a production-ready AI System-on-Module (SoM) that delivers server-class Xavier SoC performance in a form factor smaller than a credit card — 69.6 × 45 mm. Available in \u003cstrong\u003e8GB and 16GB LPDDR4x configurations\u003c\/strong\u003e, it connects to any Xavier NX-compatible carrier board via the standard 260-pin SO-DIMM interface, giving embedded engineers, OEMs, and research teams a proven path to full-AI edge systems with up to 21 TOPS, dual NVDLA engines, hardware-accelerated video encode\/decode, and multi-camera sensor fusion.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e21 TOPS AI Performance at the Edge\u003c\/strong\u003e — The Xavier NX delivers 21 TOPS of accelerated computing at 15W or 20W, and up to 14 TOPS at 10W — enabling real-time parallel neural network inference across object detection, segmentation, and video analytics entirely without cloud dependency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNVIDIA Volta GPU with 48 Tensor Cores\u003c\/strong\u003e — 384 CUDA cores and 48 Tensor Cores on the Volta architecture provide hardware-accelerated CUDA inference and TensorRT-optimised throughput on INT8 and FP16 workloads — purpose-built for production AI pipelines.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e6-Core Carmel ARM CPU\u003c\/strong\u003e — The 6-core NVIDIA Carmel ARMv8.2 64-bit CPU with 6MB L2 + 4MB L3 cache runs Linux OS management, peripheral I\/O, and application logic in parallel with GPU inference — without resource contention or scheduling delays.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDual NVDLA Deep Learning Accelerators\u003c\/strong\u003e — Two dedicated NVDLA engines offload specific inference workloads entirely from the GPU, enabling simultaneous multi-model execution where one network runs on NVDLA while another occupies the Volta GPU — maximising hardware utilisation per watt.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHardware Video Encode \u0026amp; Decode\u003c\/strong\u003e — Dedicated NVENC and NVDEC engines handle up to 2× 4K60 H.265 encoding and 2× 8K30 decoding simultaneously — freeing CPU and GPU entirely for AI inference while multi-stream video pipelines run in dedicated silicon.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMulti-Camera CSI Pipeline\u003c\/strong\u003e — Up to 6 simultaneous cameras (expandable to 24 via virtual channels) over 14-lane MIPI CSI-2 with D-PHY 1.2 at up to 30 Gbps aggregate bandwidth — purpose-built for surround-view robotics, multi-sensor inspection, and autonomous machine vision.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHigh-Speed PCIe Gen4 Expansion\u003c\/strong\u003e — One PCIe Gen3 ×1 plus one PCIe Gen4 ×4 lane (144 GT\/s total) allow direct attachment of NVMe SSDs, Wi-Fi 6 adapters, FPGA co-processors, and AI accelerator cards with no USB overhead or bandwidth bottleneck.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eOnboard 16GB eMMC 5.1 Storage\u003c\/strong\u003e — Built-in eMMC means the module boots without any external SSD or SD card — simplifying BOM, reducing carrier board complexity, and enabling clean production deployments with a single solid-state storage solution.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable Power Envelopes\u003c\/strong\u003e — Selectable 10W, 15W, and 20W power modes let you balance peak AI throughput against thermal budget — critical for passively cooled enclosures, battery-powered field systems, and fanless industrial designs.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFull JetPack \u0026amp; CUDA Ecosystem\u003c\/strong\u003e — Production-ready support for CUDA, TensorRT, cuDNN, PyTorch, TensorFlow, and NVIDIA JetPack SDK means existing AI models deploy with minimal porting effort — cloud-native container support further accelerates edge deployment pipelines.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eXavier NX 8GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eXavier NX 16GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e21 TOPS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e384-core NVIDIA Volta GPU — 48 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e6-core NVIDIA Carmel ARMv8.2 64-bit — 6MB L2 + 4MB L3\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR4x — 59.7 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e16GB 128-bit LPDDR4x — 59.7 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e16GB eMMC 5.1 (onboard)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× NVDLA Engines\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e7-Way VLIW Vision Processor\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e10W | 15W | 20W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1×1 PCIe Gen3 + 1×4 PCIe Gen4 — 144 GT\/s total\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCSI Camera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eUp to 6 cameras (24 via virtual channels) — 14-lane MIPI CSI-2, D-PHY 1.2 up to 30 Gbps\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× 4K60 | 4× 4K30 | 10× 1080p60 | 22× 1080p30 (H.265)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× 8K30 | 6× 4K60 | 12× 4K30 | 22× 1080p60 | 44× 1080p30 (H.265)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× multi-mode DP 1.4 \/ eDP 1.4 \/ HDMI 2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e10\/100\/1000 BASE-T Ethernet\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMechanical\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e69.6 × 45 mm — 260-pin SO-DIMM connector\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich Jetson Xavier NX Is Right for You?\u003c\/h3\u003e\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eBoth Xavier NX variants are built on the same Xavier SoC and deliver identical 21 TOPS AI performance — the decision is purely about memory headroom for your neural network pipeline. The 8GB module comfortably handles single or dual-model workloads and most standard embedded AI deployments. Choose the 16GB module when running large transformer models, concurrent multi-model inference, or high-resolution multi-stream video analytics where 8GB becomes the bottleneck — memory is soldered and cannot be upgraded after purchase.\u003c\/p\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eFeature\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eXavier NX 8GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eXavier NX 16GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eRAM\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8GB 128-bit LPDDR4x\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e16GB 128-bit LPDDR4x\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory Bandwidth\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e59.7 GB\/s (identical across both variants)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e21 TOPS (identical across both variants)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU \u0026amp; CPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eIdentical across both variants\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower Range\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e10W | 15W | 20W (identical across both variants)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eBest For\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSingle\/dual-model inference, standard multi-camera deployments, most embedded AI applications\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eLarge models, multi-model concurrent inference, 4K multi-stream analytics, future-proofed OEM designs\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eCommercial Robotics \u0026amp; Autonomous Machines\u003c\/strong\u003e — Run simultaneous perception, planning, and navigation models on a single Xavier NX module using the Volta GPU, dual NVDLA engines, and multi-camera CSI pipeline — purpose-built for the demanding real-time workloads of mobile robots and autonomous guided vehicles.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutomated Optical Inspection (AOI)\u003c\/strong\u003e — High-throughput defect detection on industrial production lines benefits directly from hardware-accelerated 6-camera CSI capture, TensorRT inference on Volta, and dedicated NVDLA offload — delivering sub-millisecond decision latency without cloud round-trips.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical Instruments \u0026amp; Diagnostic Imaging\u003c\/strong\u003e — Certified production-ready hardware with defined power envelopes and consistent compute performance makes the Xavier NX suitable for AI-assisted imaging devices, surgical robotics, and bedside diagnostic tools requiring regulatory-grade reliability.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Cameras \u0026amp; Intelligent Vision Systems\u003c\/strong\u003e — Embedded AI directly in the camera node eliminates central compute bottlenecks — the Xavier NX processes inference, encodes video in hardware, and communicates results over Ethernet or PCIe in a single compact thermal envelope.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHigh-Resolution Multi-Sensor Fusion\u003c\/strong\u003e — Up to 6 simultaneous high-resolution CSI cameras, dual NVDLA engines, and the Volta GPU enable complex multi-modal sensor fusion tasks in autonomous vehicles, drones, and advanced driver-assistance systems (ADAS).\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Factory \u0026amp; Industry 4.0 Systems\u003c\/strong\u003e — Deploy predictive maintenance, worker safety monitoring, and real-time quality control AI directly on the factory floor using the Xavier NX's sealed SO-DIMM form factor, fanless-friendly power modes, and industrial I\/O interfaces.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge Video Analytics at Scale\u003c\/strong\u003e — Simultaneously decode up to 44× 1080p30 streams in hardware and run AI inference without GPU involvement in decoding — making the Xavier NX ideal for large-scale smart city, retail, and security monitoring deployments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrone \u0026amp; UAV Payloads\u003c\/strong\u003e — The 10W low-power mode and compact 69.6 × 45 mm footprint enable integration into weight-critical aerial platforms for real-time aerial reconnaissance, infrastructure inspection, and autonomous mapping with fully onboard AI processing.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAIoT Gateway \u0026amp; Edge Inference Nodes\u003c\/strong\u003e — Cloud-native container support and GbE networking make the Xavier NX an efficient AIoT hub — running inference locally, aggregating sensor data, and selectively pushing processed results to cloud dashboards with minimal bandwidth consumption.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAI Research \u0026amp; Prototyping\u003c\/strong\u003e — Universities and R\u0026amp;D teams use the Xavier NX as a production-representative edge AI platform — its full CUDA, TensorRT, and JetPack ecosystem mirrors the toolchain used in large-scale deployments, making benchmarks directly transferable to product.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003e1× NVIDIA Jetson Xavier NX Module (SoM) — your selected configuration (8GB or 16GB)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: a carrier board, heatsink, thermal pad, power supply, and any cameras, display adapters, or expansion cards are not included with the module. These accessories are sold separately. A compatible carrier board and appropriate power supply are required before the module can be used.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat carrier boards are compatible with the Jetson Xavier NX Module?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Xavier NX uses a standard \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e260-pin SO-DIMM connector\u003c\/span\u003e and is compatible with any carrier board explicitly designed for the Xavier NX family — including the official NVIDIA Jetson Xavier NX Developer Kit carrier and third-party boards from Connect Tech, Auvidea, Seeed Studio, and Forecr. Critically, the Xavier NX is \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003enot electrically compatible with the original Jetson Nano carrier board\u003c\/span\u003e despite sharing the same physical SO-DIMM form factor — the two platforms use different pin assignments and cannot be interchanged. Always confirm that the carrier board vendor explicitly lists Jetson Xavier NX support before purchasing. Third-party carriers often add ruggedised connectors, multiple GbE ports, M.2 NVMe slots, and extended temperature ratings not found on the reference design.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat power supply does the Jetson Xavier NX require?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003ePower is delivered through the carrier board, so the exact supply specification is carrier-dependent. The official NVIDIA Xavier NX Developer Kit carrier uses a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e9V–20V DC barrel jack\u003c\/span\u003e (centre-positive, 5.5mm\/2.5mm), with a 19V adapter typically recommended to cover the module plus attached peripherals. The module supports three configurable power modes — \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e10W, 15W, and 20W\u003c\/span\u003e — and your supply must sustain the peak wattage of your chosen mode plus headroom for cameras, SSDs, and USB devices. For the 20W mode under full AI inference load, a supply rated at 40W or more is advisable to avoid voltage droop and instability. Always consult your specific carrier board's datasheet for exact input voltage range and connector requirements.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhich operating system and JetPack version should I use?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Xavier NX officially supports \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack 4.x and JetPack 5.x\u003c\/span\u003e, with JetPack 5.1.x being the current long-term supported release for the Xavier series, based on \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 20.04 LTS\u003c\/span\u003e. JetPack 5 ships with CUDA 11.4, TensorRT 8.x, cuDNN 8.x, OpenCV, VPI, and the Multimedia API pre-installed and ready to use. Note that \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack 6 is not supported on Xavier NX\u003c\/span\u003e — it requires the Orin-generation SoC. NVIDIA's SDK Manager tool on a Linux Ubuntu host handles flashing, firmware updates, and SDK package installation in a guided workflow. For new projects, JetPack 5.1.x is the recommended baseline.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available — does it need an SSD to boot?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eUnlike the Jetson Orin Nano, the Jetson Xavier NX includes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e16GB of onboard eMMC 5.1 storage\u003c\/span\u003e — the module can boot directly from internal flash without any external SSD or microSD card. This simplifies initial setup and reduces BOM cost for production designs. For applications requiring more capacity or faster I\/O, most carrier boards expose an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eM.2 M-key NVMe SSD slot via PCIe Gen3\u003c\/span\u003e — this is preferred for large model storage, datasets, and logging. A 128GB or 256GB NVMe SSD is a common production pairing. The onboard eMMC can simultaneously host the OS while the NVMe SSD is used exclusively for data and AI model storage.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat additional hardware do I need to get started?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAt minimum you need: a compatible \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eXavier NX carrier board\u003c\/span\u003e, a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eheatsink with thermal interface material\u003c\/span\u003e (active cooling recommended for sustained 15W\/20W loads), and a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003epower supply\u003c\/span\u003e matched to your carrier. Because the Xavier NX has onboard eMMC, you do not need to purchase a boot SSD for initial bring-up — though additional storage is recommended for production workloads. For first-time setup, a display (HDMI or DP adapter), USB keyboard, and mouse are useful — though headless SSH setup is also viable after the first flash. A host PC running \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 20.04 or 22.04 with NVIDIA SDK Manager\u003c\/span\u003e is required to flash JetPack onto the module for the first time.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson Xavier NX compare to the Jetson Orin NX?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Orin NX\u003c\/span\u003e is the next-generation successor and delivers a significant performance step up — up to 100 TOPS (Orin NX 16GB) versus 21 TOPS on the Xavier NX. The Orin NX uses the newer \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eAmpere GPU architecture\u003c\/span\u003e (vs. Volta on Xavier NX), supports JetPack 6 on Ubuntu 22.04, and offers higher memory bandwidth via LPDDR5. However, the Jetson Xavier NX remains a strong production-proven platform with a mature software ecosystem, a large installed carrier board base, and critically — \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ehardware video encoding (NVENC)\u003c\/span\u003e which the Jetson Orin Nano lacks entirely. The Xavier NX is the right choice for projects already aligned to JetPack 5 or requiring the specific 21 TOPS Volta performance envelope with full hardware encode\/decode.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO and serial interface pins are available?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson Xavier NX exposes a comprehensive I\/O set through its 260-pin SO-DIMM connector, including \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUART, SPI, I2C, I2S, CAN bus interfaces, GPIO, and PWM outputs\u003c\/span\u003e. The exact signals accessible to your application depend on your carrier board design — the official NVIDIA Developer Kit routes a subset to a 40-pin expansion header with a layout broadly familiar to Raspberry Pi users. The CAN bus interface is particularly valuable for robotics and automotive applications requiring deterministic communication with motor controllers and sensor arrays. For custom carrier board designs, NVIDIA's Xavier NX pinmux spreadsheet fully documents every signal assignment and alternate function available on the SO-DIMM connector.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs this module suitable for beginners, or is it aimed at professionals?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe standalone Jetson Xavier NX Module (SoM) is primarily aimed at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eembedded engineers, OEM product designers, and experienced AI developers\u003c\/span\u003e building custom hardware around the module. If you are new to Jetson and want to experiment first, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Xavier NX Developer Kit\u003c\/span\u003e — which bundles a carrier board — is the recommended starting point. Once comfortable with JetPack, TensorRT model deployment, and system bring-up workflows, the standalone module is the correct choice for custom carrier board designs and production scaling. NVIDIA's \"Hello AI World\" tutorial series and the active Jetson Developer Forum community substantially lower the learning curve regardless of prior Jetson experience.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat is the most common mistake users make when setting up the Xavier NX?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most frequent mistake is attempting to use a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Nano carrier board\u003c\/span\u003e with the Xavier NX module — despite the same physical SO-DIMM form factor, the two platforms are electrically incompatible and the module will not function correctly on a Nano carrier. A second common error is selecting a power supply sized only for the module's nominal wattage without accounting for attached peripherals — under-powered supplies cause random reboots and instability under AI inference load. Finally, many users overlook that \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003einitial JetPack flashing requires a Linux Ubuntu host PC\u003c\/span\u003e running NVIDIA SDK Manager — this step cannot be performed from a Windows machine directly, so plan your bring-up environment before the hardware arrives.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware updates, and community support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAll official documentation, hardware design guides, datasheets, pinmux tools, and JetPack firmware are hosted on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Developer Zone\u003c\/span\u003e at developer.nvidia.com\/embedded. NVIDIA's \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eSDK Manager\u003c\/span\u003e tool manages JetPack flashing and software updates from a Ubuntu host machine. Community support, model optimisation discussions, hardware bring-up guidance, and carrier board recommendations are available on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums — Jetson \u0026amp; Embedded Systems\u003c\/span\u003e section, actively monitored by NVIDIA engineers and the wider Jetson community. For AI deployment tutorials, NVIDIA's \"Hello AI World\" and Jetson AI Fundamentals courses are specifically designed around the Jetson Xavier hardware platform and are the recommended starting point for new users.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"8GB","offer_id":43036785901673,"sku":"NVD-005","price":62999.99,"currency_code":"INR","in_stock":true},{"title":"16GB","offer_id":43036785934441,"sku":"NVD-006","price":82149.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/Image1_720x_b3e99765-4eab-405b-bc1c-821b1e1b868c.webp?v=1774263461"},{"product_id":"nvidia-jetson-agx-orin-module","title":"NVIDIA Jetson AGX Orin Module","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;\"\u003eNVIDIA Jetson AGX Orin™ Module — Up to 275 TOPS AI Performance — Ampere Architecture — 32GB \u0026amp; 64GB LPDDR5\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe \u003cstrong\u003eNVIDIA Jetson AGX Orin™\u003c\/strong\u003e is the highest-performance module in the Jetson Orin family, delivering up to \u003cstrong\u003e275 TOPS\u003c\/strong\u003e of INT8 AI compute in a compact 100 × 87 mm form factor pin-compatible with Jetson AGX Xavier. Available in 32GB and 64GB LPDDR5 configurations, it is purpose-built for robotics, autonomous machines, and industrial edge AI at the highest performance tier.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eUp to 275 TOPS AI Performance\u003c\/strong\u003e — Delivers more than 8× the AI compute of Jetson AGX Xavier, enabling the most demanding real-time inference workloads at the edge without any cloud dependency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNVIDIA Ampere Architecture GPU\u003c\/strong\u003e — Up to 2048 CUDA cores and 64 Tensor Cores at 1.3 GHz, enabling concurrent execution of deep learning, computer vision, and graphics pipelines in a single module.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDual NVDLA v2.0 Deep Learning Accelerators\u003c\/strong\u003e — Two dedicated hardware engines offload neural network inference from the GPU, increasing throughput and power efficiency for always-on inference tasks.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHigh-Bandwidth LPDDR5 Memory\u003c\/strong\u003e — Up to 64GB of unified 256-bit LPDDR5 at 204.8 GB\/s provides ample headroom for large model weights and high-resolution sensor data without memory bottlenecks.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMulti-Camera \u0026amp; Multi-Sensor Input\u003c\/strong\u003e — Supports up to 6 simultaneous cameras (16 via virtual channels) over 16 MIPI CSI-2 lanes with D-PHY 2.1 \/ C-PHY 2.0, ideal for 360° perception systems.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePCIe Gen4 High-Speed I\/O\u003c\/strong\u003e — Up to 2×x8, 1×x4, and 2×x1 PCIe Gen4 lanes enable direct attachment of NVMe storage, FPGAs, accelerator cards, and high-speed radio modules.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHardware Video Engine\u003c\/strong\u003e — Dedicated encode and decode engines handle up to 2×4K60 encode and 3×4K60 decode (64GB) with H.265, H.264, AV1, and VP9, freeing the GPU entirely for inference.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFull NVIDIA AI Software Stack\u003c\/strong\u003e — Ships with JetPack SDK and is compatible with Isaac, DeepStream, Riva, TAO Toolkit, and Omniverse Replicator — all optimised for Orin's hardware accelerators.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable Power Envelope\u003c\/strong\u003e — Scalable from 15W to 60W (64GB) lets you tune the performance-power trade-off for battery-operated or thermally constrained deployments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrop-In Xavier Compatibility\u003c\/strong\u003e — The 699-pin Molex Mirror Mezz connector and 100 × 87 mm footprint are pin-compatible with Jetson AGX Xavier, simplifying hardware migration for existing product designs.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eJetson AGX Orin™ 32GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eJetson AGX Orin™ 64GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e200 TOPS (INT8)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e275 TOPS (INT8)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1792-core NVIDIA Ampere GPU with 56 Tensor Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2048-core NVIDIA Ampere GPU with 64 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU Max Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e930 MHz\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.3 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8-core Arm® Cortex®-A78AE v8.2 64-bit (2MB L2 + 4MB L3)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e12-core Arm® Cortex®-A78AE v8.2 64-bit (3MB L2 + 6MB L3)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU Max Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2.2 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32GB 256-bit LPDDR5 — 204.8 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64GB 256-bit LPDDR5 — 204.8 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e64GB eMMC 5.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× NVDLA v2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDLA Max Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.4 GHz\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.6 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× PVA v2\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCSI Camera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eUp to 6 cameras (16 via virtual channels) — 16 MIPI CSI-2 lanes — D-PHY 2.1 \/ C-PHY 2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×4K60, 3×4K30, 6×1080p60, 12×1080p30 (H.265, H.264, AV1)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2×4K60, 4×4K30, 8×1080p60, 16×1080p30 (H.265, H.264, AV1)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×8K30, 2×4K60, 4×4K30, 9×1080p60, 18×1080p30 (H.265, H.264, VP9, AV1)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×8K30, 3×4K60, 7×4K30, 11×1080p60, 22×1080p30 (H.265, H.264, VP9, AV1)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eUp to 2×x8, 1×x4, 2×x1 PCIe Gen4 (Root Port \u0026amp; Endpoint)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e3× USB 3.2 Gen2 (10 Gbps), 4× USB 2.0\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× GbE, 1× 10GbE\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e1× 8K60 multi-mode DP 1.4a \/ eDP 1.4a \/ HDMI 2.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOther I\/O\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e4× UART, 3× SPI, 4× I2S, 8× I2C, 2× CAN, PWM, DMIC \u0026amp; DSPK, GPIOs\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e15W – 40W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e15W – 60W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eForm Factor\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e100mm × 87mm — 699-pin Molex Mirror Mezz Connector\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eThermal\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eIntegrated Thermal Transfer Plate\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich AGX Orin Module Is Right for You?\u003c\/h3\u003e\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eBoth modules share the same PCB footprint, connector, and I\/O configuration — the choice comes down to AI throughput, CPU core count, and memory capacity. The 32GB is well-suited for most single-domain robotics workloads, while the 64GB unlocks simultaneous multi-domain inference, larger transformer models, and higher-resolution multi-camera video pipelines.\u003c\/p\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eCriteria\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eJetson AGX Orin™ 32GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eJetson AGX Orin™ 64GB\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e200 TOPS\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e275 TOPS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU Cores \/ Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1792 cores @ 930 MHz\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2048 cores @ 1.3 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU Cores\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8-core Arm A78AE\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e12-core Arm A78AE\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32GB LPDDR5\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64GB LPDDR5\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMax Power\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e40W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e60W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode (max)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1×4K60 \/ 12×1080p30\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2×4K60 \/ 16×1080p30\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eBest For\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSingle-domain robotics, AMRs, industrial inspection\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMulti-domain AI, large model inference, autonomous vehicles, multi-camera analytics\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Mobile Robots (AMRs)\u003c\/strong\u003e — Simultaneous navigation, obstacle avoidance, and object detection pipelines run concurrently using the GPU and DLA engines, enabling real-time operation without cloud dependence.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e3D Perception \u0026amp; LiDAR Fusion\u003c\/strong\u003e — High memory bandwidth and the PVA vision accelerator handle dense point clouds and multi-sensor fusion for depth estimation and scene reconstruction at robotics frame rates.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Quality Inspection\u003c\/strong\u003e — Multi-camera input and NVDLA accelerators power high-throughput defect detection on production lines, replacing multiple dedicated vision processors with a single module.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMulti-Channel Video Analytics\u003c\/strong\u003e — Hardware decoders handle up to 22 simultaneous 1080p30 streams (64GB), enabling large-scale intelligent camera networks with per-stream AI analytics via DeepStream.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eNatural Language Processing at the Edge\u003c\/strong\u003e — Sufficient compute and memory for running mid-size transformer models and large language model inference locally, enabling voice assistants and conversational AI without cloud latency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Vehicles \u0026amp; Drones\u003c\/strong\u003e — Combines real-time video encode, multi-sensor fusion, and path planning in a power-constrained form factor for next-generation UAVs and ground vehicles.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical \u0026amp; Surgical Robotics\u003c\/strong\u003e — Deterministic AI performance and broad I\/O connectivity support image-guided intervention systems, endoscopy AI, and real-time surgical assistance applications.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Retail \u0026amp; Occupancy Analytics\u003c\/strong\u003e — DeepStream-based multi-camera pipelines perform crowd analytics, shelf monitoring, and checkout automation without streaming video off-device.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSynthetic Data Generation\u003c\/strong\u003e — Omniverse Replicator integration lets teams generate labelled training datasets directly on Jetson hardware, reducing cloud data pipeline complexity for model development.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge AI Inference Server\u003c\/strong\u003e — PCIe Gen4 and 10GbE allow the module to serve multiple upstream devices as a centralised on-premise inference endpoint, replacing rack-mounted GPU servers in constrained environments.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003e1× NVIDIA Jetson AGX Orin™ Module (32GB or 64GB, as ordered)\u003c\/li\u003e\n  \u003cli\u003eIntegrated Thermal Transfer Plate (factory-attached to module)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: a carrier board is required to use this module and is not included. Accessories such as power supplies, cables, cases, and storage devices are sold separately and not included unless stated above.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs the Jetson AGX Orin module compatible with existing Jetson AGX Xavier carrier boards?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eYes. The AGX Orin uses the same \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e699-pin Molex Mirror Mezz connector\u003c\/span\u003e and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e100 × 87 mm footprint\u003c\/span\u003e as the Jetson AGX Xavier, making it a drop-in module replacement on most existing AGX Xavier carrier boards. You should confirm with your carrier board manufacturer that power delivery and BSP support have been validated for AGX Orin, as some older designs may require a firmware update or minor hardware revision. NVIDIA's official Developer Kit carrier board is fully validated for both generations out of the box.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat power supply does the Jetson AGX Orin module require?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe AGX Orin module receives power through its carrier board rather than a direct onboard connector. The 32GB variant operates between \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e15W and 40W\u003c\/span\u003e, while the 64GB variant can draw up to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e60W\u003c\/span\u003e. When using the official Jetson AGX Orin Developer Kit carrier board, a compatible DC power supply (typically 19V input, rated for your TDP tier) is required and sold separately. Always consult your carrier board documentation for the exact connector type, voltage, and current rating. Running the module at lower power modes (e.g. 15W) is useful for thermally constrained or battery-powered enclosures.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat operating system and software stack does the Jetson AGX Orin support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Orin runs \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu Linux\u003c\/span\u003e via NVIDIA's \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack SDK\u003c\/span\u003e, which bundles the BSP, CUDA, cuDNN, TensorRT, and multimedia APIs into a single flashable image. JetPack 6.x (based on Ubuntu 22.04) is the current recommended release for AGX Orin. The full NVIDIA AI software stack — including Isaac SDK, DeepStream, Riva, TAO Toolkit, and Triton Inference Server — is available and optimised for Orin's hardware accelerators. Windows and macOS are not supported; the module is Linux-only.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available, and can I add an NVMe SSD?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eBoth AGX Orin modules include \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e64GB eMMC 5.1\u003c\/span\u003e soldered on-module for the root filesystem. For expanded storage, the carrier board's PCIe Gen4 interface supports high-speed \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSDs via M.2 Key M\u003c\/span\u003e — available on the official Developer Kit carrier board. You can configure the system to boot directly from NVMe if your carrier board and BSP support it, which is strongly recommended for workloads involving high-throughput data logging or large datasets. The eMMC is sufficient for the OS and applications, but external NVMe significantly extends storage capacity and write endurance.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat accessories do I need to start using the Jetson AGX Orin module?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAt minimum you need a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ecompatible carrier board\u003c\/span\u003e, a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eDC power supply\u003c\/span\u003e rated for your TDP tier, and either a monitor or SSH access for initial setup. For the quickest start, NVIDIA's Jetson AGX Orin Developer Kit includes a full-featured carrier board with all standard ports exposed. Additional recommended accessories include a USB keyboard and mouse, a DisplayPort or HDMI monitor, an NVMe SSD for expanded storage, and a USB-A to Micro-B cable for flashing via NVIDIA SDK Manager. Camera modules, sensors, and enclosures are application-specific and sold separately.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson AGX Orin compare to the Jetson AGX Xavier?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe AGX Orin delivers up to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e8× more AI performance\u003c\/span\u003e than the Jetson AGX Xavier (275 TOPS vs approximately 32 TOPS), thanks to the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Ampere GPU architecture\u003c\/span\u003e, upgraded NVDLA v2.0 accelerators, a faster Arm Cortex-A78AE CPU cluster, and LPDDR5 memory with higher bandwidth. It adds AV1 codec support, C-PHY 2.0 CSI, and PCIe Gen4 — all absent on Xavier. Critically, it retains the same connector and form factor, making hardware migration straightforward for existing Xavier-based product designs.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO, UART, I2C, SPI, and CAN interfaces does the module expose?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eVia the 699-pin connector, the AGX Orin exposes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e4× UART, 3× SPI, 4× I2S, 8× I2C, 2× CAN, PWM, DMIC, DSPK, and multiple GPIOs\u003c\/span\u003e — the exact signals available at your system depend on how your carrier board routes them. Compared to the smaller Orin NX series, the AGX Orin provides substantially richer peripheral connectivity for complex embedded designs with multiple sensors and actuators. The 16-lane MIPI CSI-2 interface supports both D-PHY 2.1 and C-PHY 2.0 physical layers for maximum camera flexibility.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs this module suitable for beginners, or is it aimed at experienced embedded developers?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe AGX Orin module is primarily aimed at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eprofessional and industrial developers\u003c\/span\u003e building production-grade embedded AI systems. It does not include a carrier board, display, or power supply, requiring knowledge of embedded Linux, BSP flashing via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e, and hardware bring-up. Beginners are better served by the Jetson AGX Orin Developer Kit or the Jetson Orin Nano Super Developer Kit, which include carrier boards and are ready to use out of the box. The module format is intended for integration into custom hardware products.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat is the most common mistake when integrating the AGX Orin into a custom carrier board?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most frequent issue is under-specifying the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003epower delivery circuitry\u003c\/span\u003e on the carrier board. The AGX Orin can draw up to 60W at peak, and instantaneous transient currents during GPU burst workloads can spike significantly higher — insufficient bulk capacitance or undersized power rails cause instability and unexpected reboots. A second common pitfall is neglecting to apply a proper \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ethermal interface material (TIM)\u003c\/span\u003e between the module's Thermal Transfer Plate and the system heatsink or chassis. NVIDIA's carrier board design guide and the AGX Orin module datasheet both provide detailed power sequencing and thermal guidelines that must be followed precisely during hardware design.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware updates, and community support for the Jetson AGX Orin?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eAll official documentation — including the module datasheet, design guide, JetPack release notes, and BSP source — is available at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003edeveloper.nvidia.com\/embedded\/jetson-agx-orin\u003c\/span\u003e. Firmware and JetPack SDK updates are distributed through \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e and the NVIDIA L4T (Linux for Tegra) apt repository. The NVIDIA Developer Forums at forums.developer.nvidia.com host an active Jetson community for technical support, driver patches, and community-contributed projects. The jetson-inference GitHub repository maintained by Dusty Franklin provides practical inference containers and demos optimised for the Orin platform.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"32GB","offer_id":43037092184169,"sku":"NVD-012","price":140749.99,"currency_code":"INR","in_stock":true},{"title":"64GB","offer_id":43037092216937,"sku":"NVD-013","price":247799.99,"currency_code":"INR","in_stock":true},{"title":"Industrial","offer_id":43062001303657,"sku":"NVD-014","price":354899.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/51f01800-21b5-4b9e-9cad-afcba4258b97.jpg?v=1774265778"},{"product_id":"nvidia-jetson-agx-thor-developer-kit","title":"NVIDIA Jetson AGX Thor Developer Kit","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;\"\u003eNVIDIA Jetson AGX Thor Developer Kit — 2070 FP4 TFLOPS — 128 GB LPDDR5X — Blackwell GPU\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe NVIDIA® Jetson AGX Thor™ Developer Kit is the most powerful Jetson platform ever built, delivering \u003cstrong\u003e2070 FP4 TFLOPS\u003c\/strong\u003e of AI compute and 128 GB of unified memory within a 130 W power envelope. Powered by the \u003cstrong\u003eNVIDIA Blackwell GPU\u003c\/strong\u003e architecture, it achieves up to 7.5× the AI performance and 3.5× the energy efficiency of Jetson AGX Orin — the definitive platform for humanoid robotics, physical AI, and real-time generative AI at the edge.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e2070 FP4 TFLOPS AI Performance\u003c\/strong\u003e — Delivers 7.5× the AI compute of Jetson AGX Orin, enabling on-device execution of large generative AI models — including Vision-Language-Action models, LLMs, and diffusion networks — without cloud dependency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eBlackwell GPU with Multi-Instance GPU (MIG)\u003c\/strong\u003e — The 2560-core Blackwell GPU with 5th-gen Tensor Cores and MIG technology partitions compute resources across parallel inference pipelines, guaranteeing deterministic latency for simultaneous robotics workloads.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e128 GB Unified LPDDR5X Memory at 273 GB\/s\u003c\/strong\u003e — The largest memory pool in any Jetson platform eliminates bottlenecks for multi-modal AI models, giving perception, planning, and control workloads room to coexist on a single device.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e14-Core Arm® Neoverse®-V3AE CPU at 2.6 GHz\u003c\/strong\u003e — A high-performance CPU cluster handles real-time control loops, sensor preprocessing, and OS tasks in parallel with GPU inference — no external compute nodes required.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e3rd-Gen Programmable Vision Accelerator (PVA v3)\u003c\/strong\u003e — Dedicated vision hardware offloads stereo depth estimation, optical flow, and feature detection from the GPU, freeing Blackwell cores entirely for AI inference while maintaining sub-millisecond perception latency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e100 Gbps QSFP28 Networking\u003c\/strong\u003e — The onboard 4× 25GbE QSFP28 port delivers up to 100 Gbps aggregate bandwidth for high-speed sensor fusion, multi-camera streaming, and tethered robot communication — no external network card required.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePCIe Gen 5 Expansion\u003c\/strong\u003e — M.2 Key M (×4 PCIe Gen5) and M.2 Key E (×1 PCIe Gen5) slots provide next-generation storage and wireless throughput, future-proofing the platform for high-bandwidth peripherals.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eScalable 40–130 W Power Envelope\u003c\/strong\u003e — Configurable TDP from 40 W to 130 W lets you balance performance and power for mobile, tethered, or lab deployments — all from a single developer kit form factor.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHumanoid Robot-Ready Design\u003c\/strong\u003e — Engineered for seamless integration with existing humanoid robot platforms, featuring CAN headers, JTAG, automation headers, and a compact 243 × 112 × 57 mm chassis for rapid tethered prototyping.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eComplete NVIDIA AI Software Stack\u003c\/strong\u003e — Ships with JetPack 7, NVIDIA Isaac for robotics, Holoscan for sensor processing, and Metropolis for visual AI agents — the entire ecosystem for physical AI development in one kit.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eDetails\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2070 TFLOPS (FP4 Sparse) \/ 1035 TFLOPS (FP8)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2560-core NVIDIA Blackwell architecture GPU with 5th-gen Tensor Cores, Multi-Instance GPU (MIG) with 10 TPCs\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU Max Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1.57 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e14-core Arm® Neoverse®-V3AE 64-bit CPU, 1 MB L2 cache per core, 16 MB shared L3 cache\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU Max Frequency\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2.6 GHz\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× PVA v3 (3rd-Generation Programmable Vision Accelerator)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e128 GB 256-bit LPDDR5X at 273 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1 TB NVMe (M.2 Key M slot, PCIe Gen5 ×4)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× NVENC\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× NVDEC\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCamera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eHigh-speed camera via QSFP slot, USB camera support\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eM.2 Key M (×4 PCIe Gen5), M.2 Key E (×1 PCIe Gen5)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× USB-A (3.2 Gen2), 2× USB-C (3.1)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× 5GbE RJ45, 1× QSFP28 (4× 25GbE, up to 100 Gbps aggregate)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× HDMI 2.0b, 1× DisplayPort 1.4a\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOther I\/O\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eQSFP connector, M.2 Key E \u0026amp; M slots, CAN headers, automation headers, LED, JTAG, fan connector, audio header, power jack, RTC connector\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e40 W – 130 W (configurable TDP modes)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMechanical\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e243.19 mm × 112.40 mm × 56.88 mm\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eThermal\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eThermal Transfer Plate (TTP), optional fan or heatsink\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eHumanoid Robotics\u003c\/strong\u003e — The 128 GB memory pool and 2070 TFLOPS enable on-device execution of Vision-Language-Action (VLA) models, giving humanoid robots real-time perception, language-driven reasoning, and dexterous manipulation capabilities without cloud latency.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Mobile Robots (AMR)\u003c\/strong\u003e — MIG-isolated inference pipelines and 100 Gbps QSFP28 networking allow AMRs to simultaneously fuse LiDAR, radar, and multi-camera data at production throughput, enabling reliable navigation in complex, dynamic environments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSurgical \u0026amp; Medical Robotics\u003c\/strong\u003e — Deterministic low-latency inference on the Blackwell GPU combined with real-time CAN bus control enables surgical robots to respond to force feedback and visual cues within milliseconds of each other.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Inspection \u0026amp; Quality Control\u003c\/strong\u003e — PVA v3 and dual NVENC\/NVDEC engines enable parallel high-resolution visual inspection across production lines simultaneously, detecting sub-millimeter defects at line speed without sacrificing throughput.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eEdge Generative AI (LLMs \u0026amp; VLMs)\u003c\/strong\u003e — 2070 FP4 TFLOPS and 128 GB of unified memory make it feasible to run 70B-parameter language and vision models entirely on-device, powering embodied AI agents that reason and plan without cloud round-trips.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrone \u0026amp; UAV Perception\u003c\/strong\u003e — The 40 W minimum power mode and compact 243 × 112 × 57 mm chassis allow power-constrained aerial platforms to run full AI inference on multi-camera payloads without prohibitive weight or battery penalties.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eReal-Time Video Analytics\u003c\/strong\u003e — Dual NVENC and NVDEC hardware engines enable simultaneous encoding, decoding, and AI analysis of multiple 4K video streams for smart city surveillance, retail analytics, and live sports applications.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart Manufacturing \u0026amp; Collaborative Robots\u003c\/strong\u003e — NVIDIA Holoscan sensor processing pipelines and real-time CAN control allow cobots to dynamically adapt to unstructured environments alongside human workers, enabling safe and productive collaboration.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDigital Twins \u0026amp; Multi-Sensor Fusion\u003c\/strong\u003e — The QSFP28 and 5GbE networking, CAN headers, and 128 GB memory allow a single Jetson AGX Thor to aggregate and synchronise data from dozens of heterogeneous sensors in real time, powering accurate digital twin pipelines.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eResearch \u0026amp; Sim-to-Real Transfer\u003c\/strong\u003e — PCIe Gen 5 expansion and high-bandwidth networking allow researchers to connect GPU clusters or high-speed storage arrays, making the kit ideal for dataset collection, model fine-tuning, and validating simulated AI behaviours on physical hardware.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003eNVIDIA Jetson T500 Module\u003c\/li\u003e\n  \u003cli\u003eReference Carrier Board\u003c\/li\u003e\n  \u003cli\u003eHeatsink and Fan\u003c\/li\u003e\n  \u003cli\u003e1 TB NVMe SSD (pre-installed in M.2 Key M slot)\u003c\/li\u003e\n  \u003cli\u003e140 W DC Power Supply\u003c\/li\u003e\n  \u003cli\u003e802.11ax (Wi-Fi 6) Wireless NIC (pre-installed in M.2 Key E slot)\u003c\/li\u003e\n  \u003cli\u003eQuick Start Guide\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: accessories such as QSFP cables, USB peripherals, displays, cameras, and cases are sold separately and not included unless stated above.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat operating systems and software frameworks are compatible with the Jetson AGX Thor?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Thor runs \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 22.04 LTS\u003c\/span\u003e via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack 7\u003c\/span\u003e, NVIDIA's official SDK for the Jetson platform. It is fully compatible with the NVIDIA AI software stack, including \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Isaac\u003c\/span\u003e for robotics, \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eHoloscan\u003c\/span\u003e for sensor processing, and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eMetropolis\u003c\/span\u003e for visual AI. Popular deep learning frameworks including PyTorch, TensorFlow, ONNX Runtime, and TensorRT are all fully supported on the Aarch64 platform. Containerised workflows via Docker and the NVIDIA NGC catalogue are natively supported, enabling rapid deployment of pre-optimised AI models without manual dependency management.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat are the power requirements for the Jetson AGX Thor Developer Kit?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe developer kit is powered by the included \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e140 W DC power supply\u003c\/span\u003e via a barrel jack connector on the reference carrier board. The module itself operates across a configurable TDP range of \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e40 W to 130 W\u003c\/span\u003e, selectable through software power modes within JetPack. The 40 W mode suits power-constrained or mobile deployments, while 130 W mode unlocks the full 2070 FP4 TFLOPS of AI compute for maximum throughput. Always use the supplied or a compatible 140 W+ adapter — underpowering under heavy inference loads can cause thermal throttling or unexpected shutdowns.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhich OS and firmware version does the Jetson AGX Thor ship with?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Thor ships with \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetPack 7\u003c\/span\u003e, NVIDIA's latest SDK built specifically for the Blackwell architecture and generative AI workloads. JetPack 7 bundles the Linux kernel, board support package (BSP), CUDA, cuDNN, TensorRT, and the full suite of NVIDIA developer libraries in a single install. Firmware and OS updates are distributed via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e (a free tool) or directly through the Jetson APT software repository. It is strongly recommended to flash the latest stable JetPack 7 release from the NVIDIA developer portal before beginning development to ensure up-to-date drivers, security patches, and library versions.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available, and can I upgrade the pre-installed drive?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe developer kit ships with a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e1 TB NVMe SSD\u003c\/span\u003e pre-installed in the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eM.2 Key M slot\u003c\/span\u003e (PCIe Gen5 ×4), providing fast sequential throughput for large model files and datasets. The slot accepts standard 2280 or 2242 form-factor NVMe drives, so upgrading to a larger capacity is straightforward. There is no soldered eMMC on this module — the NVMe SSD is both the boot and primary storage device. External USB storage via the USB-A and USB-C ports is also supported for dataset staging, backup, or additional working space.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat additional accessories do I need to get started with the Jetson AGX Thor?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eTo begin developing you will need a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003emonitor with HDMI or DisplayPort\u003c\/span\u003e, a USB keyboard, and a USB mouse — none of which are included. A \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eQSFP28 cable or transceiver\u003c\/span\u003e is required to use the 100 Gbps networking port, and a standard Ethernet cable connects the 5GbE RJ45 port. The included 802.11ax Wi-Fi NIC means wireless internet is available out of the box for basic connectivity. For initial flashing, a Linux host PC with \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e (free download) is required to write the JetPack image to the NVMe drive over USB-C.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson AGX Thor compare to the Jetson AGX Orin?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Thor is a generational leap over the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson AGX Orin\u003c\/span\u003e, delivering up to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e7.5× higher AI compute\u003c\/span\u003e (2070 FP4 TFLOPS vs ~275 TOPS) and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e3.5× better energy efficiency\u003c\/span\u003e. Memory doubles from a maximum of 64 GB on Orin to 128 GB on Thor, while memory bandwidth increases from 204.8 GB\/s to 273 GB\/s. The GPU architecture upgrades from Ampere to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eBlackwell\u003c\/span\u003e — adding 5th-gen Tensor Cores and full MIG support — while the CPU moves from Cortex-A78AE to the higher-performance \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eArm Neoverse-V3AE\u003c\/span\u003e. Connectivity also advances with PCIe Gen 5 (up from Gen 4) and the addition of a QSFP28 100 Gbps network port absent on Orin.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat GPIO, CAN, and I\/O interfaces are available for robotics integration?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe reference carrier board exposes a broad set of robotics-focused interfaces including \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eCAN bus headers\u003c\/span\u003e for industrial motor controllers and actuators, \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eautomation headers\u003c\/span\u003e for digital I\/O integration, and a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJTAG\u003c\/span\u003e port for hardware-level debugging and bring-up. Additional I\/O includes an audio header, LED connector, fan connector for active cooling, and an RTC connector for real-time clock battery backup. USB connectivity is provided via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e2× USB-A 3.2 Gen2\u003c\/span\u003e and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e2× USB-C 3.1\u003c\/span\u003e ports, with simultaneous HDMI 2.0b and DisplayPort 1.4a display outputs also available.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs the Jetson AGX Thor suitable for beginners, or is it aimed at advanced developers?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Thor is primarily designed for \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eadvanced robotics engineers and AI researchers\u003c\/span\u003e who require maximum on-device compute for generative AI and physical AI applications. That said, JetPack 7 provides a familiar Ubuntu-based environment, and NVIDIA's extensive tutorials, sample code, and pre-built Docker containers significantly lower the entry barrier. Developers with solid Python, Linux, and machine learning fundamentals can get productive quickly using \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eIsaac ROS\u003c\/span\u003e and the NVIDIA Hello AI World guide. For those entirely new to embedded AI, starting on a Jetson Orin Nano or Jetson AGX Orin to build foundational skills before stepping up to the Thor is a practical approach.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat is the most common mistake developers make when first setting up the Jetson AGX Thor?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most common mistake is running the kit at full 130 W performance mode without adequate airflow or without the fan installed — this causes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ethermal throttling\u003c\/span\u003e within minutes of a sustained heavy inference workload. Always verify the heatsink is properly seated and the fan connector is plugged in before enabling high-power modes. A second frequent error is installing \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eCUDA\u003c\/span\u003e, cuDNN, or TensorRT from upstream x86 binaries — Aarch64 Jetson builds must come from the JetPack APT repository or NVIDIA SDK Manager, as generic x86 packages will not run on the Arm CPU. Always use the JetPack-native library sources to avoid subtle incompatibilities that can be difficult to diagnose.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find documentation, community support, and firmware updates for the Jetson AGX Thor?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eOfficial documentation, JetPack SDK downloads, and software release notes are available on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer website\u003c\/span\u003e at developer.nvidia.com\/embedded. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums\u003c\/span\u003e (forums.developer.nvidia.com) host an active Jetson community with dedicated AGX Thor threads where NVIDIA engineers regularly respond to technical questions. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA NGC catalogue\u003c\/span\u003e (ngc.nvidia.com) provides pre-optimised AI model containers, reference applications, and ready-to-deploy pipelines for JetPack 7. For robotics-specific resources, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eIsaac ROS GitHub repository\u003c\/span\u003e and NVIDIA Isaac documentation site offer tutorials, ROS 2 packages, and worked examples for the most common physical AI use cases.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":43037233643625,"sku":"NVD-016","price":394299.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/484d577f-6619-4f51-9441-0bc68e70fd96.jpg?v=1774268105"},{"product_id":"nvidia-jetson-agx-orin-developer-kit64gb","title":"NVIDIA JETSON AGX ORIN DEVELOPER KIT(64GB)","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;\"\u003eNVIDIA Jetson AGX Orin Developer Kit — 275 TOPS Edge AI — 64 GB LPDDR5 — 2048-Core Ampere GPU\u003c\/h2\u003e\n\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe \u003cstrong\u003eNVIDIA Jetson AGX Orin Developer Kit\u003c\/strong\u003e is NVIDIA's most powerful edge AI platform, delivering up to \u003cstrong\u003e275 TOPS of AI performance\u003c\/strong\u003e — up to 8× more than Jetson AGX Xavier — from a compact module that shares the same form factor and software stack across the entire Orin family. Powered by a 12-core Arm Cortex-A78AE CPU, 2048-core Ampere GPU, dual NVDLA v2.0 deep-learning accelerators, and 64 GB of LPDDR5 unified memory, it is built for robotics, autonomous machines, industrial vision, and on-device generative AI inference backed by the full JetPack 6 SDK.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e275 TOPS AI Performance\u003c\/strong\u003e — Delivers up to 8× more INT8 throughput than the Jetson AGX Xavier, enabling real-time inference on large transformer, vision, and diffusion models at the edge without cloud dependency.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e2048-Core NVIDIA Ampere GPU with 64 Tensor Cores\u003c\/strong\u003e — Full CUDA 12 support accelerates TensorRT deployment, mixed-precision workloads, and sparse tensor operations in a single unified memory space shared with the CPU.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e64 GB 256-bit LPDDR5 — 204.8 GB\/s Bandwidth\u003c\/strong\u003e — Shared GPU–CPU memory eliminates data-copy overhead, making it practical to run multi-billion-parameter models and large vision pipelines without batching constraints.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDual NVDLA v2.0 + PVA v2.0 Accelerators\u003c\/strong\u003e — Dedicated silicon for DNN inference and computer-vision pre\/post-processing frees GPU cycles for the most latency-sensitive tasks in the pipeline.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMulti-Camera \u0026amp; High-Throughput Video Pipeline\u003c\/strong\u003e — 16-lane MIPI CSI-2 input with hardware encode up to 4K60 and decode up to 8K30 supports multi-stream analytics and spatial AI systems with minimal CPU overhead.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable 15–60 W TDP with Four Power Modes\u003c\/strong\u003e — Software-selectable 15 W, 30 W, 50 W, and 60 W MAXN modes let you tune performance vs. power budget without changing hardware — ideal for both lab and deployed environments.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePCIe Gen4, 10 GbE, Wi-Fi 5, Bluetooth 5.0 \u0026amp; Rich Expansion I\/O\u003c\/strong\u003e — x16 PCIe Gen4 slot, M.2 Key M (x4 PCIe Gen4 NVMe), M.2 Key E, 10 Gigabit Ethernet, 6× USB 3.2 ports, DisplayPort 1.4a, and a 40-pin GPIO header cover virtually every sensor or peripheral integration scenario.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eFull NVIDIA JetPack 6 SDK Ecosystem\u003c\/strong\u003e — Ships ready to flash with JetPack 6.x (CUDA 12.6, TensorRT 10.3, cuDNN 9.3, VPI 3.2), Isaac ROS, TAO Toolkit, and DeepStream — the same APIs used in NVIDIA data-centre systems, so models move from cloud to edge without rewriting inference code.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n    \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eDetails\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUp to 275 TOPS\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNVIDIA Ampere — 2048 CUDA cores, 64 Tensor Cores\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e12-core Arm Cortex-A78AE v8.2 64-bit @ 2.2 GHz (3 MB L2 + 6 MB L3)\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× NVDLA v2.0\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePVA v2.0\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB 256-bit LPDDR5 — 204.8 GB\/s bandwidth\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB eMMC 5.1 + M.2 Key M (x4 PCIe Gen4) NVMe expansion\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× 4K60 | 4× 4K30 | 8× 1080p60 | 16× 1080p30 (H.265)\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1× 8K30 | 3× 4K60 | 6× 4K30 | 12× 1080p60 | 24× 1080p30 (H.265)\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCamera\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e16-lane MIPI CSI-2 connector\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10 GbE (RJ45) + Wi-Fi 802.11ac + Bluetooth 5.0 (pre-installed M.2 module)\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePCIe\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ex16 slot (x8 PCIe Gen4) + M.2 Key M (x4 PCIe Gen4) + M.2 Key E (x1 PCIe Gen4)\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUSB\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× USB 3.2 Gen2 Type-C | 2× USB 3.2 Gen2 Type-A | 2× USB 3.2 Gen1 Type-A | 1× USB 2.0 Micro-B\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplayPort 1.4a with MST support\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPIO \/ Expansion\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e40-pin header (I2C, GPIO, SPI, CAN, I2S, UART, DMIC), 12-pin automation header, 10-pin audio header, 10-pin JTAG, 4-pin fan, 2-pin RTC\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e15 W \/ 30 W \/ 50 W \/ 60 W MAXN (software-configurable), DC barrel jack 7–20 V\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDimensions\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e110 mm × 110 mm × 71.65 mm\u003c\/td\u003e\n        \u003c\/tr\u003e\n    \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich Jetson Orin Is Right for You?\u003c\/h3\u003e\n\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eThe Jetson Orin family scales from 67 TOPS up to 275 TOPS — all sharing the same Ampere GPU architecture, JetPack SDK, and software pipeline. Choose your module based on AI workload size, memory footprint, and power budget; the same Isaac ROS and TensorRT code runs on every variant without modification.\u003c\/p\u003e\n\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n    \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:600px;border:0;\"\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eModel\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAI Performance\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eGPU Cores\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eCPU Cores\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eMemory\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003ePower\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:700;word-wrap:break-word;color:#BAFF02;\"\u003eJetson AGX Orin Dev Kit\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:700;word-wrap:break-word;color:#BAFF02;\"\u003e275 TOPS\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2048 Ampere\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e12-core A78AE\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB LPDDR5\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e15–60 W\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eJetson AGX Orin 32GB\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e200 TOPS\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2048 Ampere\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8-core A78AE\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32 GB LPDDR5\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e15–40 W\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eJetson Orin NX 16GB\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e157 TOPS\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1792 Ampere\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8-core A78AE\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e16 GB LPDDR5\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10–40 W\u003c\/td\u003e\n        \u003c\/tr\u003e\n        \u003ctr\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eJetson Orin Nano 8GB\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e67 TOPS\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e1024 Ampere\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e6-core A78AE\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8 GB LPDDR5\u003c\/td\u003e\n            \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e7–25 W\u003c\/td\u003e\n        \u003c\/tr\u003e\n    \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Mobile Robots (AMR)\u003c\/strong\u003e — Run simultaneous SLAM, multi-sensor fusion, path planning, and obstacle avoidance in real time using all compute engines on a single board, replacing entire racks of compute in earlier robot architectures.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Machine Vision \u0026amp; Quality Control\u003c\/strong\u003e — Process up to 16 simultaneous 1080p camera feeds with hardware decode and run defect-detection CNNs via NVDLA at throughputs that match high-speed production-line rates.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart City \u0026amp; Traffic Analytics\u003c\/strong\u003e — Deploy multi-stream 4K video analytics for vehicle counting, licence-plate recognition, and anomaly detection at the edge without transmitting raw video to the cloud.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical Imaging AI\u003c\/strong\u003e — Perform on-device inference for CT, MRI, and ultrasound analysis with the large unified memory pool enabling full-resolution 3D model inference that GPU-constrained devices cannot fit in memory.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Drones \u0026amp; UAVs\u003c\/strong\u003e — Execute low-latency real-time perception, obstacle avoidance, and landing-zone detection without cloud round-trips, with configurable TDP keeping airborne power budgets manageable.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eGenerative AI \u0026amp; LLM Inference at the Edge\u003c\/strong\u003e — Run quantised large language models and diffusion models locally with TensorRT-LLM, enabling private, on-premises AI assistants and content generation without sending sensitive data off-device.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAgricultural Robotics\u003c\/strong\u003e — Power multi-spectral crop-monitoring drones and harvesting robots that need vision, navigation, and actuation compute in a package capable of operating in field conditions.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eRetail \u0026amp; In-Store Analytics\u003c\/strong\u003e — Run anonymous people-counting, queue-length detection, and shelf-inventory AI simultaneously across multiple camera feeds without transmitting footage off-site.\u003c\/li\u003e\n    \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eRobotics Research \u0026amp; Rapid Prototyping\u003c\/strong\u003e — The full set of carrier board interfaces — PCIe, GPIO, MIPI CSI, USB, CAN, I2S — lets research teams attach virtually any sensor or actuator and iterate without custom PCB work, with Isaac ROS providing ready-made perception primitives.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n    \u003cli\u003eJetson AGX Orin module (64 GB) with heatsink and fan on reference carrier board\u003c\/li\u003e\n    \u003cli\u003eAzureWave AW-CB375NF Wi-Fi 802.11ac + Bluetooth 5.0 M.2 module (pre-installed)\u003c\/li\u003e\n    \u003cli\u003e90 W USB PD Type-C power supply with regional cords (US, EU, UK)\u003c\/li\u003e\n    \u003cli\u003eUSB Type-A to Type-C cable\u003c\/li\u003e\n    \u003cli\u003eQuick Start Guide\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: accessories such as NVMe SSDs, MIPI cameras, display adapters, and microSD cards are sold separately and not included unless stated above.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat software and ML frameworks are compatible with the Jetson AGX Orin Developer Kit?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Orin runs the full \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack 6 SDK\u003c\/span\u003e stack, which includes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eCUDA 12.6\u003c\/span\u003e, \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eTensorRT 10.3\u003c\/span\u003e, cuDNN 9.3, OpenCV, and VPI 3.2. It is fully compatible with \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eIsaac ROS\u003c\/span\u003e, DeepStream SDK, NVIDIA TAO Toolkit, and ROS 2 (Humble). Popular ML frameworks — PyTorch, TensorFlow, ONNX Runtime, and Triton Inference Server — all have validated ARM64 builds for Jetson. Because the Orin shares the same CUDA and TensorRT API versions as NVIDIA data-centre GPUs, models trained and optimised in the cloud can be deployed to the edge with minimal code changes.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat are the power requirements for the Jetson AGX Orin Developer Kit?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe developer kit is powered by the included \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e90 W USB Power Delivery Type-C adapter\u003c\/span\u003e, which connects to the USB-C port on the carrier board. The module supports four software-selectable power modes: \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e15 W, 30 W, 50 W, and 60 W (MAXN)\u003c\/span\u003e, selectable via the nvpmodel utility or jtop. An alternative \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eDC barrel jack (7–20 V input)\u003c\/span\u003e is also present on the carrier board for integration into custom power systems. When building a production enclosure, account for peak wattage plus any peripherals connected to the PCIe slot or USB ports when sizing your power supply.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat operating system does the Jetson AGX Orin run?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Orin runs \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson Linux\u003c\/span\u003e (formerly L4T — Linux for Tegra), an Ubuntu-based distribution currently built on \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUbuntu 22.04 LTS\u003c\/span\u003e with Linux kernel 5.15. You flash it using \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e from a host PC running Ubuntu 20.04 or 22.04 — the developer kit does not ship pre-flashed. JetPack 6.x is recommended for all new projects; JetPack 5.x (Ubuntu 20.04 base) remains available for projects requiring legacy compatibility. Red Hat Device Edge is also tested and supported as an alternative OS for enterprise deployments.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options does the Jetson AGX Orin Developer Kit support?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe module includes \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e64 GB eMMC 5.1\u003c\/span\u003e on-module flash for the operating system and core applications. For expanded storage, the carrier board provides an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eM.2 Key M slot (x4 PCIe Gen4)\u003c\/span\u003e supporting NVMe SSDs in the 2280 form factor — this is the recommended path for large datasets and model files, as Gen4 NVMe delivers far higher sequential throughput than the eMMC. A \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003emicroSD slot (UHS-1, SDR104 mode)\u003c\/span\u003e is also available for removable media. NVMe drives are not included in the box and are sold separately; a PCIe Gen4 NVMe 2280 SSD is strongly recommended for any serious inference or training workload.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat accessories do I need to get started with the Jetson AGX Orin Developer Kit?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eOut of the box you need: a display connected via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eDisplayPort 1.4a\u003c\/span\u003e (or a DP-to-HDMI adapter, sold separately), a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eUSB keyboard and mouse\u003c\/span\u003e, and a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ehost PC running Ubuntu 20.04 or 22.04\u003c\/span\u003e for flashing JetPack via NVIDIA SDK Manager. The 90 W power supply and USB-C cable are included. For camera-based projects, USB cameras work immediately out of the box; MIPI CSI-2 cameras require an optional camera adapter board. An NVMe SSD (sold separately) is strongly recommended for working with large models or datasets beyond what the eMMC comfortably accommodates.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson AGX Orin compare to the Jetson AGX Xavier?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Orin delivers \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eup to 8× more AI compute\u003c\/span\u003e than the AGX Xavier — 275 TOPS vs. 30 TOPS. The GPU steps from 512-core Volta to \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e2048-core Ampere\u003c\/span\u003e with Tensor Cores and sparse-tensor support not available on Xavier. The CPU moves from 8-core Carmel to a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e12-core Cortex-A78AE at 2.2 GHz\u003c\/span\u003e, approximately 1.7× faster per-thread. Memory doubles to 64 GB LPDDR5 with 1.4× higher bandwidth than Xavier's LPDDR4x. Both modules share the same 699-pin SOM connector, so many Xavier-compatible carrier boards are physically compatible with Orin — though software must be updated to JetPack 5 or 6.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO pins and interfaces does the 40-pin header provide?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe standard \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e40-pin expansion header\u003c\/span\u003e (2×20, 2.54 mm pitch, Raspberry Pi HAT-compatible pinout) provides I2C (×2), SPI (×2), UART (×2), I2S, CAN, GPIO, PWM, 3.3 V, and 5 V supply rails — exact pin allocation depends on the device-tree overlay loaded at boot. Beyond the 40-pin header, the carrier board adds a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e12-pin automation header\u003c\/span\u003e (SPI, DMIC, GPIO), a 10-pin audio header, a 10-pin JTAG debug header, a 4-pin fan header, and a 2-pin RTC battery connector. This combination lets you attach motor drivers, IMUs, CAN bus networks, and audio peripherals simultaneously without an I\/O hub.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs the Jetson AGX Orin Developer Kit suitable for beginners or only advanced users?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe developer kit targets professional developers, researchers, and advanced students — it is not a general-purpose single-board computer. Familiarity with Linux CLI, Python or C++, and a working knowledge of ML frameworks is expected. That said, NVIDIA provides free \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson AI Courses\u003c\/span\u003e on the NVIDIA Deep Learning Institute (DLI) covering JetPack setup, TensorRT optimisation, and Isaac ROS basics. The official \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eHello AI World\u003c\/span\u003e tutorial gets a real-time object-detection pipeline running in under an hour. Developers comfortable with Raspberry Pi who want to move into serious AI inference will find the learning curve worthwhile but should budget time for the initial JetPack flashing and SDK environment setup.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat is the most common mistake when setting up the Jetson AGX Orin for the first time?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most frequent pitfall is attempting to flash JetPack from a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eWindows or macOS host\u003c\/span\u003e — NVIDIA SDK Manager only runs on x86_64 Ubuntu 20.04 or 22.04. A second common issue is forgetting to enter \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eForce Recovery mode\u003c\/span\u003e before connecting to the host PC — hold the Force Recovery button while pressing Power; the board will not appear as a USB device otherwise. Finally, many users do not expand the root partition after flashing and run out of space when installing JetPack components on the 64 GB eMMC — installing immediately to an \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVMe SSD\u003c\/span\u003e or running resize2fs on the eMMC partition right after first boot eliminates this issue entirely.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n    \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find documentation, community support, and firmware updates for the Jetson AGX Orin?\u003c\/p\u003e\n    \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eOfficial documentation lives on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer portal\u003c\/span\u003e (developer.nvidia.com\/embedded), including the Jetson AGX Orin User Guide, JetPack release notes, and the Jetson Linux driver package. Firmware and JetPack updates are distributed via \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e (GUI installer) or the Jetson Linux archive for command-line flashing. Community support is active on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums\u003c\/span\u003e under the Jetson \u0026amp; Embedded Systems category — most hardware and software questions have existing threads. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ejetson-containers\u003c\/span\u003e project on GitHub provides a curated library of ready-to-run Docker containers for LLMs, diffusion models, vision pipelines, and robotics workloads, saving significant environment-setup time.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":43037640130665,"sku":"NVD-015","price":225299.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/878f723d-be71-4a0d-9a3d-d37c1bb38709.jpg?v=1774270036"},{"product_id":"nvidia-dgx-spark","title":"NVIDIA DGX Spark","description":"\u003cdiv class=\"nv-title text h--medium aem-GridColumn--tablet--12 aem-GridColumn--offset--tablet--0 aem-GridColumn--default--none aem-GridColumn--phone--none aem-GridColumn--phone--12 aem-GridColumn--tablet--none aem-GridColumn aem-GridColumn--default--7 aem-GridColumn--offset--phone--0 aem-GridColumn--offset--default--0\"\u003e\n\u003cdiv id=\"nv-title-10385987dd\" class=\"general-container-text\"\u003e\n\u003cdiv class=\"text-left lap-text-left tab-text-center mob-text-center\"\u003e\n\u003ch2 class=\"title\"\u003eSpark Something Big\u003c\/h2\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"nv-text text aem-GridColumn--tablet--12 aem-GridColumn--offset--tablet--0 aem-GridColumn--default--none aem-GridColumn--phone--none aem-GridColumn--phone--12 aem-GridColumn--tablet--none aem-GridColumn aem-GridColumn--default--7 aem-GridColumn--offset--phone--0 aem-GridColumn--offset--default--0\"\u003e\n\u003cdiv id=\"nv-text-ee7fc092f1\" class=\"general-container-text\"\u003e\n\u003cdiv class=\"text-left lap-text-left tab-text-center mob-text-center\"\u003e\n\u003cdiv class=\"description\"\u003e\n\u003cp\u003e\u003cspan class=\"p--large\"\u003ePowered by the NVIDIA GB10 Grace Blackwell Superchip, NVIDIA DGX Spark™ delivers up to one petaFLOP1 of FP4 AI performance in a power-efficient, compact form factor. With a preinstalled NVIDIA AI software stack and 128 GB of memory, developers can prototype, fine-tune, and deploy the latest reasoning AI models. Additionally, NVIDIA NemoClaw, part of the NVIDIA Agent Toolkit is an open source agent development platform for building, evaluating, and optimizing safer, long-running autonomous agents directly from the desktop.\u003c\/span\u003e\u003c\/p\u003e\n\u003ch2 class=\"title\"\u003eFeatures\u003c\/h2\u003e\n\u003cp\u003e\u003cimg\u003e\u003cimg\u003e\u003cimg src=\"https:\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/Screenshot_2026-03-25_195424.png?v=1774448704\" alt=\"\"\u003e\u003c\/p\u003e\n\u003ch2\u003eA Grace Blackwell AI supercomputer on Your Desk\u003c\/h2\u003e\n\u003ch4\u003e1. Based on NVIDIA Grace Blackwell Architecture\u003c\/h4\u003e\n\u003cp\u003eAt the heart of DGX Spark is the new GB10 Grace Blackwell Superchip, based on the Grace Blackwell architecture and optimized for a desktop form factor. GB10 features a powerful Blackwell GPU with fifth-generation Tensor Cores and FP4 support, delivering up to 1000 AI TOPS of compute. GB10 also includes a high-performance Grace 20-core Arm CPU to supercharge data preprocessing and orchestration, speeding up model tuning and real-time inferencing. The GB10 Superchip uses the NVLink™-C2C to deliver a CPU+GPU coherent memory model with 5X the bandwidth of PCIe Gen 5.\u003c\/p\u003e\n\u003ch4\u003e2. Work With the Latest Generation of Large-Parameter Generative AI Models\u003c\/h4\u003e\n\u003cp\u003eWith 128 GB of unified system memory and support for the FP4 data format, DGX Spark can support AI models of up to 200B parameters, enabling AI developers to prototype, fine-tune and inference the latest generation of AI reasoning models—such as DeepSeek R1 distilled versions up to 70 billion parameters—on their desktop. With built-in NVIDIA ConnectX™ network technology, two DGX Spark systems can be connected to work on even larger models such as Llama 3.1 405B.\u003c\/p\u003e\n\u003ch4\u003e3. Develop Locally, Deploy Anywhere at Scale\u003c\/h4\u003e\n\u003cp\u003eDGX Spark provides developers with a powerful, experimentation ground for prototyping models and AI applications, freeing up valuable compute resources in their cluster environments better suited for training and deploying production models. Leveraging the NVIDIA AI platform software architecture makes it possible for DGX Spark users to seamlessly move their models from their desktop to DGX Cloud or any accelerated cloud or data center infrastructure with virtually no code changes, making it easier than ever to prototype, fine-tune, and iterate.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eTechnical Specifications:\u003c\/strong\u003e\u003c\/p\u003e\n\u003ctable style=\"width: 100.036%; height: 490px;\"\u003e\n\u003ctbody\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eArchitecture\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eNVIDIA Grace Blackwell\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eGPU\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eNVIDIA Blackwell Architecture\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eCPU\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e20 core Arm (10 Cortex-X925 + 10 Cortex-A725)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eCUDA Cores\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eNVIDIA Blackwell Generation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eTensor Cores\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e5th Generation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eRT Cores\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e4th Generation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eTensor Performance\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e1 PFLOP\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eSystem Memory\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e128 GB LPDDR5x, coherent unified system memory\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eMemory Interface\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e256-bit\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eMemory Bandwidth\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eUp to 273 GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eStorage\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e4 TB NVMe M.2 with self-encryption\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eUSB\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e4x USB Type-C\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eEthernet\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e1x RJ-45 connector, 10 GbE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eNIC\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eConnectX-7 NIC @ 200 Gbps\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eWi-Fi\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eWiFi 7\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eBluetooth\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eBT 5.4 with LE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eAudio Output\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eHDMI multichannel audio output\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003ePower Supply\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e240 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eGB10 TDP\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e140 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eDisplay Connectors\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e1x HDMI 2.1a\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eNVENC \/ NVDEC\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e1x \/ 1x\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eOS\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003eNVIDIA DGX™ OS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eSystem Dimensions\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e150 mm (L) x 150 mm (W) x 50.5 mm (H)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"height: 19.6px;\"\u003e\n\u003ctd style=\"width: 27.3396%; height: 19.6px;\"\u003eSystem Weight\u003c\/td\u003e\n\u003ctd style=\"width: 70.2005%; height: 19.6px;\"\u003e1.2 kg\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003cp\u003e \u003c\/p\u003e\nDatasheet: \u003ca href=\"https:\/\/nvdam.widen.net\/s\/tlzm8smqjx\/workstation-datasheet-dgx-spark-gtc25-spring-nvidia-us-3716899-web\" rel=\"noopener\" target=\"_blank\"\u003e\u003cspan style=\"text-decoration: underline; color: rgb(43, 0, 255);\"\u003e\u003cstrong\u003eDownload\u003c\/strong\u003e\u003c\/span\u003e\u003c\/a\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e","brand":"Nvidia","offers":[{"title":"Default Title","offer_id":43059701940329,"sku":"NVD-017","price":499999.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/spark-3qtr-right.png?v=1774449134"},{"product_id":"nvidia-jetson-agx-xavier-module","title":"Nvidia Jetson AGX Xavier Module","description":"\u003ch2 style=\"font-size:1.4em;font-weight:700;margin:0 0 12px;line-height:1.4;\"\u003eNVIDIA Jetson AGX Xavier — 32 TOPS AI Inference — 512-Core Volta GPU — Industrial Edge AI Module\u003c\/h2\u003e\n\u003cp style=\"margin:0 0 20px;line-height:1.7;color:#e0e0e0;\"\u003eThe NVIDIA Jetson AGX Xavier delivers workstation-class AI compute in a palm-sized 100×87 mm system-on-module, making fully autonomous machines practical at the edge. Available in \u003cstrong\u003e64GB\u003c\/strong\u003e, \u003cstrong\u003e32GB\u003c\/strong\u003e, and \u003cstrong\u003eIndustrial\u003c\/strong\u003e configurations, it scales from a power-efficient 10W mobile profile to a ruggedised 40W Industrial variant rated for -40°C to 85°C with ECC memory and a dedicated Safety Cluster Engine.\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eKey Highlights\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e32 TOPS AI Performance\u003c\/strong\u003e — Run demanding real-time inference at the edge without cloud dependency; the dual NVDLA engines plus 512-core Volta GPU work in parallel to sustain up to 32 trillion operations per second.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e512-Core Volta GPU with 64 Tensor Cores\u003c\/strong\u003e — Accelerates deep learning inference across vision, natural language, and multi-sensor fusion tasks simultaneously, using the same architecture as NVIDIA data-centre GPUs.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e2× NVDLA Deep Learning Accelerators\u003c\/strong\u003e — Dedicated fixed-function engines offload neural network inference from the GPU, cutting latency and freeing GPU cycles for graphics and general compute.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e8-Core NVIDIA Carmel ARMv8.2 CPU\u003c\/strong\u003e — 8 MB L2 + 4 MB L3 cache hierarchy sustains the multi-threaded control stacks, middleware, and OS workloads typical of production robotics systems.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e256-Bit LPDDR4x at 136.5 GB\/s\u003c\/strong\u003e — Industry-leading memory bandwidth prevents bottlenecks when multiple accelerators are streaming large activation maps and sensor data concurrently.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eUp to 36 Camera Virtual Channels\u003c\/strong\u003e — 16 MIPI CSI-2 lanes support up to 6 direct cameras or 36 virtual channels, enabling full-surround perception systems for autonomous vehicles and robotics platforms.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003e8× PCIe Gen4 High-Speed I\/O\u003c\/strong\u003e — Direct attachment of NVMe SSDs, LiDAR controllers, radar modules, and FPGAs without throughput compromise, backed by 3× USB 3.1 for peripheral connectivity.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSafety Cluster Engine (Industrial Only)\u003c\/strong\u003e — Two Arm Cortex-R5 cores operating in lockstep mode provide the functional safety supervision required for ASIL-compliant industrial and defence deployments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eConfigurable Power Profiles\u003c\/strong\u003e — Three selectable TDP modes (10W \/ 15W \/ 30W on standard variants; 20W \/ 40W on Industrial) let system designers tune compute density versus thermal budget for each deployment context.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eJetPack SDK \u0026amp; Full CUDA-X Ecosystem\u003c\/strong\u003e — Ships with TensorRT, cuDNN, DeepStream, and VisionWorks pre-optimised for the Volta architecture, enabling rapid model deployment from training to production inference.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eTechnical Specifications\u003c\/h3\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:560px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eSpecification\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier 64GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier 32GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier Industrial\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eAI Performance\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e32 TOPS\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e30 TOPS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eGPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003eNVIDIA Volta architecture — 512 CUDA Cores \u0026amp; 64 Tensor Cores\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCPU\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003e8-Core NVIDIA Carmel ARMv8.2 64-bit — 8 MB L2 + 4 MB L3\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDL Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003e2× NVDLA\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVision Accelerator\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003e2× PVA\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSafety Cluster Engine\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e—\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× Arm Cortex-R5 in lockstep\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB 256-bit LPDDR4x — 136.5 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32 GB 256-bit LPDDR4x — 136.5 GB\/s\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32 GB 256-bit LPDDR4x (ECC) — 136.5 GB\/s\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e32 GB eMMC 5.1\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB eMMC 5.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eCamera\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003eUp to 6 cameras (36 via virtual channels) — 16 lanes MIPI CSI-2 | 8 lanes SLVS-EC — D-PHY 1.2 (up to 40 Gbps) — C-PHY 1.1 (up to 62 Gbps)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUp to 6 cameras (36 via virtual channels) — 16 lanes MIPI CSI-2 — D-PHY 1.2 (up to 40 Gbps) — C-PHY 1.1 (up to 62 Gbps)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Encode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e4× 4K60 | 8× 4K30 | 16× 1080p60 | 32× 1080p30 (H.265) — 30× 1080p30 (H.264)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× 4K60 | 6× 4K30 | 12× 1080p60 | 24× 1080p30 (H.265\/H.264)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eVideo Decode\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e2× 8K30 | 6× 4K60 | 12× 4K30 | 26× 1080p60 | 52× 1080p30 (H.265) — 30× 1080p30 (H.264)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× 8K30 | 4× 4K60 | 8× 4K30 | 18× 1080p60 | 36× 1080p30 (H.265) — 24× 1080p30 (H.264)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eUPHY\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e8× PCIe Gen4 | 8× SLVS-EC — 3× USB 3.1 — Single Lane UFS\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e8× PCIe Gen4 — 3× USB 3.1 — Single Lane UFS\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e10W | 15W | 30W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e20W | 40W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eNetworking\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003e10\/100\/1000 BASE-T Ethernet\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eDisplay\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003eThree multi-mode DP 1.2 \/ eDP 1.4 \/ HDMI 2.0 a\/b\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOther I\/O\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003eUSB 2.0 — UART, SPI, CAN, I2C, I2S, DMIC \u0026amp; DSPK, GPIOs\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOperating Temperature\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e-25°C to 80°C (at TTP)\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e-40°C to 85°C (at TTP)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOperating Lifetime\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e5 Years\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10 Years\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMechanical\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"3\"\u003e100 mm × 87 mm — 699-pin connector — Integrated Thermal Transfer Plate\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhich AGX Xavier Variant Is Right for You?\u003c\/h3\u003e\n\u003cp style=\"margin:0 0 16px;line-height:1.7;color:#e0e0e0;\"\u003eAll three modules share the same Volta GPU, CPU, and accelerator core but differ in memory capacity, storage, power envelope, and environmental ratings. Choose by your memory headroom requirement and deployment environment.\u003c\/p\u003e\n\u003cdiv style=\"width:100%;overflow-x:auto;margin:0 0 24px;\"\u003e\n  \u003ctable style=\"width:100%;border-collapse:collapse;font-size:14px;min-width:460px;border:0;\"\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eFactor\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier 64GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier 32GB\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:2px solid #3a3a3a;font-weight:700;color:#BAFF02;\"\u003eAGX Xavier Industrial\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eMemory\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB LPDDR4x\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32 GB LPDDR4x\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e32 GB LPDDR4x + ECC\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStorage\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e32 GB eMMC 5.1\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e64 GB eMMC 5.1\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003ePower Profiles\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e10W \/ 15W \/ 30W\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e20W \/ 40W\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOperating Temp\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e-25°C to 80°C\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e-40°C to 85°C\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eSafety Cluster\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e—\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e2× Arm Cortex-R5 (lockstep)\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eOperating Lifetime\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\" colspan=\"2\"\u003e5 Years\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border-bottom:1px solid #3a3a3a;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003e10 Years\u003c\/td\u003e\n    \u003c\/tr\u003e\n    \u003ctr\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eBest For\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eLarge multi-model pipelines, LLM-at-edge, memory-intensive sensor fusion\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eStandard production robotics, AMR, smart cameras, research platforms\u003c\/td\u003e\n      \u003ctd style=\"padding:10px 12px;border:0;font-weight:600;word-wrap:break-word;color:#e0e0e0;\"\u003eHarsh environments, safety-critical systems, defence, aerospace, factory automation\u003c\/td\u003e\n    \u003c\/tr\u003e\n  \u003c\/table\u003e\n\u003c\/div\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eCommon Applications \u0026amp; Use Cases\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 20px;padding-left:22px;line-height:1.6;list-style-position:outside;color:#e0e0e0;\"\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Mobile Robots (AMR)\u003c\/strong\u003e — Real-time simultaneous localisation, sensor fusion, path planning, and obstacle avoidance all run concurrently on a single module, eliminating the need for a separate compute stack.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eAutonomous Ground Vehicles\u003c\/strong\u003e — High-bandwidth MIPI CSI-2 and PCIe Gen4 interfaces let the module ingest LiDAR, radar, and camera data in parallel while running full self-driving perception and planning networks.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eIndustrial Automated Inspection\u003c\/strong\u003e — Runs multiple high-resolution defect detection models simultaneously for PCB assembly, weld quality, and surface finish inspection on high-speed production lines.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eSmart City \u0026amp; Traffic Analytics\u003c\/strong\u003e — Up to 36 virtual camera channels combined with DeepStream SDK enable dense multi-camera vehicle counting, incident detection, and pedestrian flow analysis in real time.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eMedical Imaging \u0026amp; AI Diagnostics\u003c\/strong\u003e — ECC memory on the Industrial variant ensures bit-error protection for diagnostic accuracy; the module's inference throughput supports real-time radiology AI at the point of care.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDrone \u0026amp; UAV Perception\u003c\/strong\u003e — The compact 100 × 87 mm footprint and 10W minimum power mode make the Jetson AGX Xavier viable for weight- and power-constrained aerial platforms requiring onboard inference.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eLogistics \u0026amp; Warehouse Automation\u003c\/strong\u003e — Powers pick-and-place robot vision, conveyor OCR, item dimensioning, and multi-arm coordination with the deterministic latency needed for high-throughput fulfilment centres.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eDefence \u0026amp; Aerospace Systems\u003c\/strong\u003e — The Industrial variant's 50G operational shock rating, 5G RMS vibration tolerance, and -40°C to 85°C range satisfy requirements for airborne ISR platforms and field-deployed systems.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003ePrecision Agriculture \u0026amp; Field Robotics\u003c\/strong\u003e — Drives multi-spectral sensor fusion, AI weed detection, and autonomous path following on tractors and field robots operating in unstructured outdoor environments.\u003c\/li\u003e\n  \u003cli style=\"margin-bottom:14px;padding-left:0;line-height:1.6;\"\u003e\n\u003cstrong\u003eResearch \u0026amp; Rapid Prototyping\u003c\/strong\u003e — Full JetPack SDK compatibility and CUDA-X access let teams go from PyTorch or TensorFlow training to optimised TensorRT deployment without leaving the Jetson ecosystem.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 10px;color:#e0e0e0;\"\u003eWhat's in the Box\u003c\/h3\u003e\n\u003cul style=\"margin:0 0 12px;padding-left:22px;line-height:1.8;color:#e0e0e0;\"\u003e\n  \u003cli\u003e1× NVIDIA Jetson AGX Xavier Module (64GB, 32GB, or Industrial — as ordered)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp style=\"font-size:13px;margin:0 0 20px;line-height:1.6;color:#a0a0a0;\"\u003e\u003cem\u003eNote: accessories such as power supplies, cables, cases, and SD cards are sold separately and not included unless stated above.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3 style=\"font-size:1.15em;font-weight:700;margin:24px 0 16px;color:#e0e0e0;\"\u003eFrequently Asked Questions\u003c\/h3\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat operating systems are compatible with the Jetson AGX Xavier?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe Jetson AGX Xavier runs \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA JetPack SDK\u003c\/span\u003e, which is built on Ubuntu Linux (18.04 for JetPack 4.x, 20.04 for JetPack 5.x). JetPack bundles the full BSP, kernel, and all CUDA-X libraries including \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eTensorRT, cuDNN, DeepStream, and VisionWorks\u003c\/span\u003e. Custom Linux distributions are also supported provided a compatible device tree and BSP are used. Windows is not natively supported on this module.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat are the power requirements for the Jetson AGX Xavier?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe standard AGX Xavier modules (64GB and 32GB) operate across three selectable power modes: \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e10W, 15W, and 30W\u003c\/span\u003e. The Industrial variant uses \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e20W and 40W\u003c\/span\u003e profiles. Input voltage is supplied through the carrier board — the NVIDIA reference carrier board accepts 9–20V DC. The module itself requires a carrier board for power delivery and does not have a direct mains connector.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow does the Jetson AGX Xavier compare to the Jetson AGX Orin?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson AGX Orin\u003c\/span\u003e is NVIDIA's next-generation platform and delivers up to 275 TOPS — roughly 8× the AI performance of the AGX Xavier's 32 TOPS. Orin uses the Ampere GPU architecture and introduces 12 Arm Cortex-A78AE CPU cores in place of Xavier's 8-core Carmel design. For new designs prioritising peak inference throughput, Orin is the recommended platform; the AGX Xavier remains an excellent choice for production programmes where it is already qualified and cost-efficiency is the priority.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat storage options are available beyond the built-in eMMC?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eIn addition to the on-module \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e32 GB (64 GB on Industrial) eMMC 5.1\u003c\/span\u003e, the AGX Xavier supports external NVMe SSDs via its \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ePCIe Gen4 interface\u003c\/span\u003e — most carrier boards expose an M.2 Key M slot for this purpose. A Single Lane UFS interface is also available for additional flash storage. SD cards are not directly supported by the module itself, though some carrier boards add an SD slot via their own controller. NVMe is the recommended choice for high-throughput data logging and large dataset storage.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat accessories are needed to start using the Jetson AGX Xavier module?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe bare module requires a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ecompatible carrier board\u003c\/span\u003e that breaks out its 699-pin connector interfaces — the NVIDIA Jetson AGX Xavier Developer Kit carrier board is the most common starting point. You will also need a \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e9–20V DC power supply\u003c\/span\u003e, a USB-C cable for flashing, a display (HDMI or DisplayPort), and a host PC running Ubuntu with NVIDIA SDK Manager for initial setup. Camera modules, Ethernet cables, and NVMe SSDs are optional but typical for most applications.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eDoes the Jetson AGX Xavier support popular AI frameworks like PyTorch and TensorFlow?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eYes — \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ePyTorch, TensorFlow, and ONNX Runtime\u003c\/span\u003e all run on Jetson AGX Xavier via JetPack's CUDA-X libraries. Models are typically trained on a desktop or cloud GPU, then exported and optimised for inference using \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA TensorRT\u003c\/span\u003e, which can achieve 2–4× throughput improvement over unoptimised frameworks on the same hardware. The NVIDIA NGC catalogue also provides pre-trained, optimised model containers ready to deploy directly on Jetson.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eHow many GPIO pins and communication interfaces does the Jetson AGX Xavier expose?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe AGX Xavier exposes a rich I\/O set via its \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e699-pin board-to-board connector\u003c\/span\u003e, including GPIO, UART, SPI, CAN, I2C, I2S, DMIC, and DSPK interfaces. The reference carrier board breaks out a 40-pin header compatible with Raspberry Pi HATs for GPIO access. On top of this, \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003e3× USB 3.1, Gigabit Ethernet, and three display outputs\u003c\/span\u003e (DP\/eDP\/HDMI) are available. The exact pins exposed at system level depend on the carrier board design.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eIs the Jetson AGX Xavier module suitable for beginners or is it aimed at professional engineers?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe bare module is primarily aimed at \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eprofessional embedded engineers and system integrators\u003c\/span\u003e designing custom carrier boards for production hardware. Beginners and researchers are better served by the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetson AGX Xavier Developer Kit\u003c\/span\u003e, which includes the carrier board, cooling solution, and power supply needed to get started immediately. Once comfortable with the JetPack ecosystem and CUDA development, transitioning to the production module is straightforward.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 12px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhat is the most common mistake when deploying the Jetson AGX Xavier in production?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe most frequent issue is leaving the module on the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003edefault 15W power mode\u003c\/span\u003e without validating whether a higher TDP is needed — many deployments underperform simply because the power profile was never changed from factory default. A close second is skipping \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eTensorRT optimisation\u003c\/span\u003e before deployment: running unoptimised PyTorch models can yield 3–4× lower throughput than the hardware is capable of. Always benchmark with \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003ejetson_clocks\u003c\/span\u003e and a TensorRT-compiled engine before finalising your system design.\u003c\/p\u003e\n\u003c\/div\u003e\n\n\u003cdiv style=\"background:#1a1a1a;border-left:3px solid #BAFF02;border-radius:4px;padding:18px 20px;margin:0 0 4px;\"\u003e\n  \u003cp style=\"font-weight:700;color:#BAFF02;margin:0 0 10px;line-height:1.5;font-size:0.97em;\"\u003eWhere can I find official documentation, firmware updates, and community support?\u003c\/p\u003e\n  \u003cp style=\"margin:0;line-height:1.75;font-size:0.94em;color:#e0e0e0;\"\u003eThe primary resources are the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Jetson Linux Developer Guide\u003c\/span\u003e at developer.nvidia.com\/embedded, the \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA Developer Forums\u003c\/span\u003e (forums.developer.nvidia.com — Jetson \u0026amp; Embedded Systems section), and \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eNVIDIA SDK Manager\u003c\/span\u003e for downloading and flashing JetPack releases. The \u003cspan style=\"color:#BAFF02;font-weight:600;\"\u003eJetsonHacks\u003c\/span\u003e community site and GitHub organisation also offer hardware teardowns, setup scripts, and project tutorials maintained by the broader Jetson community.\u003c\/p\u003e\n\u003c\/div\u003e\n","brand":"Nvidia","offers":[{"title":"32GB","offer_id":43061978038377,"sku":"NVD-009","price":140749.99,"currency_code":"INR","in_stock":true},{"title":"Industrial","offer_id":43061978103913,"sku":"NVD-011","price":195999.99,"currency_code":"INR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/files\/buy-jetson_agx_xavier_module--XavierModule_White.jpg?v=1774506155"}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0672\/4229\/5401\/collections\/NicePng_nvidia-logo-png_1776479.png?v=1775978017","url":"https:\/\/edgetechrobotics.com\/collections\/nvidia-products.oembed","provider":"EdgeTech Robotics","version":"1.0","type":"link"}