Thorium
Accelerated by NVIDIA Jetson 
Accelerated by NVIDIA Jetson
Helium is one of the smallest NVIDIA® Jetson™-based embedded systems on the market. Despite its compact size, it delivers impressive computing power and is designed for use in extreme environments. Thanks to its IP67 rating, it is waterproof, dustproof, and shock-resistant, ensuring reliable operation even under the harshest conditions.
- NVIDIA Jetson Orin Nano / NX with Super Mode: up to 157 TOPS of AI performance
- Camera Interfaces: 4x GMSL2 with up to 20 Gbit/s
- Industrial-grade connectivity: 2x Gigabit Ethernet (M12), 1x USB-C 3.2 (10 Gbit/s)
- Rugged design: IP67 rated, waterproof, dustproof, shock-resistant
- Customizable connectivity: WiFi, GPS, 4G or 5G
Thanks to its robust design, Helium is ideally suited for use in industrial environments, the automotive sector, mining, and aerospace applications.
Thorium: Schließt die Lücke zwischen klassischen Embedded-Plattformen und servernaher KI-Rechenleistung direkt am Edge.
Thorium wurde für KI-Anwendungen entwickelt, bei denen klassische Embedded-Systeme nicht mehr ausreichen und Cloud- oder Serverlösungen technisch oder wirtschaftlich keinen Sinn ergeben. Es kombiniert extreme Rechenleistung mit industrietauglicher Mechanik und ermöglicht es, komplexe KI-Modelle direkt am Edge auszuführen – deterministisch, latenzarm und unabhängig von zentraler Infrastruktur.
- Servernahe KI-Leistung am Edge: Über 2000 FP4-TFLOPS ermöglichen den Einsatz sehr großer Modelle direkt im Feld.
- Hohe Systemintegration: Compute, Netzwerk, Storage und Security in einer kompakten, industrietauglichen Plattform.
- Für anspruchsvolle Umgebungen: IP67 und robuste Industrie-Konnektoren.
- Skalierbar & produktionsfähig: Ausgelegt für Serienprodukte und langfristige Verfügbarkeit.
Accelerated by NVIDIA Jetson
NVIDIA Jetson is the leading AI-at-the-edge computing platform with over half a million developers. With pre-trained AI models, developer SDKs and support for cloud-native technologies across the full Jetson lineup, manufacturers of intelligent machines and AI developers can build and deploy high-quality, software-defined features on embedded and edge devices targeting robotics, AIoT, smart cities, healthcare, industrial applications, and more. Cloud-native support helps manufacturers and developers implement frequent improvements, improve accuracy, and use the latest features with Jetson-based edge AI devices.
| Jetson T5000 | Jetson T4000 | |
|---|---|---|
| KI Performance | 2070 TOPS | 1200 TOPS |
| GPU | 2560-core NVIDIA Blackwell GPU with 96 Tensor Cores | 1536-core NVIDIA Blackwell GPU with 64 Tensor Cores |
| CPU | 14-core Arm® Neoverse®-V3AE 64-bit CPU | 12-core Arm® Neoverse®-V3AE 64-bit CPU |
| RAM | 128 GB 256-bit LPDDR5X | 64 GB 256-bit LPDDR5X |
| Storage | 2x NVMe M-Key Support 2280 & 2242 | |
| Power consumption | 40W - 130W | 40W - 70W |
Technical details
| Supported NVIDIA SOMs | Jetson Orin Nano or NX |
| Operating system | NVIDIA JetPack™ SDK 6.2: Ubuntu 22. 04 (Linux Kernel 5.15) |
| Power supply | 12V - 48V DC |
| Power consumption | max. 80 W |
| Cooling | Passive cooling: can be adapted by yourself or by PCB Arts Active cooling: with DIN-Rail option or Wall-Mount |
| Standard Interfaces | 4x GMSL2 (MAX96724), 1x USB-C USB 3. 2 (10Gbit/s, Screw terminal), 2x Gigabit Ethernet ports (M12) |
| Integrated features | 4x Opto Isolated IO's, IMU: ICM-42670-P, Crypto-Chip: ATSHA204A, external RTC |
| Optional Interfaces | 256 GB internal SSD, WiFi, 4G, 5G |
| Debug Header | via USB-C with PCB Arts Adapter |
| Display Output | via USB- C with PCB Arts Adapter |
| Casing | Anodised aluminium |
| Dimensions | 85 mm x 90 mm x 27 mm Height +18 mm (with active cooling solution) |
| Weight | 350gr. |
| Environmental temperature | -40 °C ... +85 °C (integrated heating) |
| IP class | IP 67 |
| Certification | CE, EMV, RoHS |
