Image similar
Image similar
Nvidia Jetson Thor T5000 128 GB 14 x 2.4 GHz
EAN:
Part number:
Item no:
Technical specifications
- AI-OUTPUT 2070 TFLOPS (FP4)
- GPU:2560 cores, Nvidia Blackwell architecture with 96 tenor cores
- CPU:14-core arm Neoverse V3AE, up to 2.6 GHz
- 128 GB LPDDR5X, 273 GB / s memory
- Video encode: Up to 6x 4Kp60 (H.265), 24x 1080p60
- Video decoding: Up to 4x 8Kp30 (H.265), 92x 1080p30
- Camera: Up to 32 cameras, MIPI CSI-2, C-PHY 2.1, D-PHY 2.1
- Network: 4x 25 GbE
- USB: 3x USB 3.2, 4x USB 2.0
- 4x HDMI 2.1/DisplayPort 1.4a
- Power consumption 40 W - 130 W.
- Dimension: 100 mm x 87 mm
- Processor cores (No.)
14 x
128 GB
- Clock speed
2.4 GHz
- Memory capacity
128 GB
- External ports
HDMI™
4x USB 2.0
- Length
100 mm
- Width
87 mm
- Product type
SBC
Documents & Downloads
User & Safety Instructions
User & Safety Instructions 3531175 Nvidia Jetson Thor T5000 128 GB 14 x 2.4 GHz
Description
The Nvidia JETSON Thor module provides maximum computing power for AI-supported robotics and physical AI applications. With up to 2070 FP4 TFLOPS AI Compute, 128 GB memory and flexible power consumption between 40 W and 130 W it enables efficient processing of complex generative AI models. The combination of GPU, CPU and extensive I/O options makes the Ideal module for real-time multi-sensor applications.
- Powerful SOM: The module is based on the Nvidia Blackwell architecture GPU with transformer engine and multi-instance GPU support.
- Flexible integration: A wide range of I/O interfaces, PCIe Gen5, USB 3.2 and support for up to 32 cameras enable a wide range of applications.
- Optimized efficiency: Up to 3.5 times higher energy efficiency compared to Nvidia AGX ORIN, suitable for mobile and stationary systems.
- Comprehensive AI software: Supports the Nvidia AI software stacks such as Isaac for robotics, Metropolis for visual AI and holoscan for sensor processing.
This text is machine translated.
- Powerful SOM: The module is based on the Nvidia Blackwell architecture GPU with transformer engine and multi-instance GPU support.
- Flexible integration: A wide range of I/O interfaces, PCIe Gen5, USB 3.2 and support for up to 32 cameras enable a wide range of applications.
- Optimized efficiency: Up to 3.5 times higher energy efficiency compared to Nvidia AGX ORIN, suitable for mobile and stationary systems.
- Comprehensive AI software: Supports the Nvidia AI software stacks such as Isaac for robotics, Metropolis for visual AI and holoscan for sensor processing.
This text is machine translated.