Details
- In accordance with applicable export control laws and regulations, NVIDIA prohibits the sale of this product and its associated kits and modules in any form to Cuba, Iran, North Korea, Syria, the Crimea region, the Donetsk and Luhansk regions of Ukraine, Russia, Belarus, and any entities on the Entity List. (Countries or regions are subject to change based on updates or revisions to export control regulations.)
- Not for use in military, aerospace, nuclear, or supercomputing-related products or applications.
- We do not provide any technical support or services to the aforementioned countries, regions, entities, or restricted end-use applications.
- This product is a precision device; once opened, it cannot be returned or exchanged unless there is a quality issue.
delivering up to 1200 / 2070 FP4 TFLOPS of AI compute
The N-VIDIA Jetson Thor series of modules gives you the ultimate platform for physical AI and robotics, delivering up to 1200 FP4 TFLOPS / 2070 FP4 TFLOPS of AI compute and 64GB / 128 GB of memory with power configurable between 40 W and 70 W, or between 40 W and 130 W. They deliver over 7.5x higher AI compute than N-VIDIA Jetson AGX Orin, with 3.5x better energy efficiency.
This system-on-module (SoM) features an N-VIDIA Blackwell architecture GPU with a Transformer engine and Multi-Instance GPU (MIG) support to effortlessly run the latest generative AI models. N-VIDIA Jetson Thor accelerates low-latency, real-time multi-sensor applications with a 12-core / 14-core Arm® Neoverse®-V3AE CPU, 3x / 4x 25 GbE networking, and extensive I/O options for sensor fusion. It also includes a suite of accelerators, including a third-generation Programmable Vision Accelerator (PVA), dual encoders and decoders, an optical flow accelerator, and more.
Jetson Thor series supports a wide range of generative AI models, from vision language action (VLA) models like N-VIDIA Isaac™ GR00T for humanoids to all popular large language models (LLMs) and vision language models (VLMs).
To deliver a seamless cloud-to-edge experience, Jetson Thor runs the N-VIDIA AI software stack for physical AI applications, including N-VIDIA Isaac for robotics, N-VIDIA Metropolis for visual agentic AI, and N-VIDIA Holoscan for sensor processing. You can also build AI agents at the edge using N-VIDIA agentic AI workflows like Video Search and Summarization (VSS).
for Advanced AI and Robotics
N-VIDIA Jetson Thor series modules give you the ultimate platform for physical AI and robotics, delivering up to 1200 FP4 TFLOPS / 2070 FP4 TFLOPS of AI compute and 64GB / 128 GB of memory with power configurable between 40 W and 70 W, or between 40 W and 130 W. They deliver higher AI compute than N-VIDIA Jetson AGX Orin.
Blackwell GPU, sensor processing, and robotic AI software stack
Unlock Physical AI with the Power of Jetson Thor
| Model |
Jetson AGX Thor Developer Kit |
Jetson Thor T5000 | Jetson Thor T4000 |
|---|---|---|---|
| AI Performance |
2070 TFLOPS (FP4 - sparse) |
1200 TFLOPS (FP4 - sparse) |
|
| GPU |
2560-core N-VIDIA Blackwell architecture GPU with 96 fifth-gen Tensor Cores, Multi-Instance GPU (MIG) with 10 TPCs |
1536-core N-VIDIA Blackwell architecture GPU with 64 fifth-gen Tensor Cores, Multi-Instance GPU (MIG) with 6 TPCs |
|
| GPU Max Frequency | 1.57 GHz | ||
| CPU |
14-core Arm® Neoverse®-V3AE 64-bit CPU, 1 MB L2 cache per core, 16 MB shared system L3 cache |
12-core Arm® Neoverse®-V3AE 64-bit CPU, 1 MB L2 cache per core, 16 MB shared system L3 cache |
|
| CPU Max Frequency | 2.6 GHz | ||
| Vision Accelerator | 1x PVA v3 | ||
| Memory |
128 GB 256-bit LPDDR5X 273 GB/s |
64 GB 256-bit LPDDR5X 273 GB/s |
|
| Storage |
1 TB NVMe M.2 Key M slot |
Supports NVMe through PCIe Supports Solid State Drives through USB3.2 |
|
| Video Encoding |
6x 4Kp60 (H.265) 12x 4Kp30 (H.265) 24x 1080p60 (H.265) 50x 1080p30 (H.265) 48x 1080p30 (H.264) 6x 4Kp60 (H.264) |
||
| Video Decoding |
4x 8Kp30 (H.265) 10x 4Kp60 (H.265) 22x 4Kp30 (H.265) 46x 1080p60 (H.265) 92x 1080p30 (H.265) 82x 1080p30 (H.264) 4x 4Kp60 (H.264) |
||
| Cameras |
HSB camera via QSFP slot USB camera |
Up to 20 cameras via HSB Up to 6 cameras through 16x lanes MIPI CSI-2 Up to 32 cameras using Virtual Channels C-PHY 2.1 (10.25 Gbps) D-PHY 2.1 (40 Gbps) |
|
| PCIe* |
M.2 Key M slot with x4 PCIe Gen5 M.2 Key E slot with x1 PCIe Gen5 |
Up to 8 lanes - Gen5 Root port only - C1 (x1) and C3 (x2) Root port or endpoint - C2 (x1), C4 (x8), and C5 (x4) |
|
| USB Ports* |
2x USB-A (3.2 Gen2) 2x USB-C (3.1) |
xHCI host controller with integrated PHY (Up to) 3x USB 3.2 4x USB 2.0 |
|
| Networking* |
1x 50GbE RJ45 connector 1x QSFP28 (4x 25 GbE) |
4x 25 GbE | 3x 25 GbE |
| Displays |
1x HDMI 2.0b 1x DisplayPort 1.4a |
4x shared HDMI 2.1 VESA DisplayPort 1.4a-HBR2, MST |
|
| Other I/O |
QSFP connector M.2 Key E expansion slot (WLAN/BT, x1 PCIe, USB2.0, UART, I2C, I2S) M.2 Key M connector (NVMe for storage) PCIe x4 lanes, I2C, PCIe x2 lanes 2x 13-pin CAN header 2x 6-pin automation header LED JTAG connector (2x 5-pin header) 1x fan connector – 12V, PWM Tach audio panel header (2x 5-pin) Microsoft power jack RTC backup battery connector (2-pin) |
5x I2S/2x Audio Hub (AHUB), 2x DMIS, 4x UART, 4x CAN, 3x SPI, 12x I2C, 6x PWM outputs | 5x I2S/2x Audio Hub (AHUB), 2x DMIS, 4x UART, 3x SPI, 13x I2C, 6x PWM outputs |
| Power | 40W ~ 130W | 40W ~ 130W | 40W ~ 70W |
| Dimensions |
243.19 x 112.40 x 56.88 (mm) Thermal Transfer Plate (TTP) and optional fan or heatsink |
100 x 87 (mm) 699 pin B2B connector Integrated Thermal Transfer Plate (TTP) with Heatpipe |
|
|
- Refer to the "Software Features" section of the latest N-VIDIA Jetson Linux Developer Guide for a list of supported features. - These specifications are preliminary and subject to change. |
|||
Quick Overview
OPTIONS Jetson Thor T4000 / T5000 x1
