NVIDIA® Jetson Thor™ Developer Kit for Physical AI and Robotics
- Powered by NVIDIA® Jetson Thor™, up to 2070 TFLOPS
- IEC 60068-2-64 and IEC 60068-2-27 certified for shock and vibration resistance
- Supports LLM and VLM workloads for real-time decision-making and interaction
- M.2 Key M 2280 slot (PCIe Gen5 x4) for high-speed data processing
- QSFP28 Quad 25GbE LAN for high-throughput sensor data
- 24/7 remote monitoring, control, and OTA deployment via Allxon
Scalable, Multi-Stream Edge AI in a Compact Fanless Form Factor
- Powered by Intel® N-series processor (Twin Lake)
- Supports M.2 AI accelerators including Hailo, Axelera, DeepX, and MemryX — up to 214 TOPS (INT8)
- Processes multiple concurrent video streams on a single device
- Fanless design with wide-temperature operation from -40°C to +70°C
- Wide-range 9–36V DC power input for flexible industrial deployment
- Dual 2.5GbE, Wi-Fi 6E, and LTE connectivity
Rugged, Fanless Embedded AI System for Industrial Deployments
- Intel® Core™ Ultra 7/5 processor (Series 1 & 2) with multi-engine AI compute via CPU, GPU (Intel® Arc), and NPU (Intel® AI Boost)
- OpenVINO™ toolkit for optimized AI inference
- Dual-channel DDR5-5600 SO-DIMM; < 96GB
- Three 2.5GbE LAN with TSN support and one GbE for reliable industrial networking
- Wide operating temperature from -40°C to +60°C and power input from 9 to 60VDC
- Flexible storage with dual front-access swappable HDD/SSD bays and one M.2 Key M 2280 slot
