Skip to product information
1 of 4

NVIDIA® Jetson AGX Thor™ Developer Kit with Blackwell GPU, 2070 FP4 TFLOPS, 14‑core CPU, 128 GB LPDDR5X, rich I/O, 130 W

NVIDIA® Jetson AGX Thor™ Developer Kit with Blackwell GPU, 2070 FP4 TFLOPS, 14‑core CPU, 128 GB LPDDR5X, rich I/O, 130 W

Seeed Studio

Regular price $4,599.00 USD
Regular price Sale price $4,599.00 USD
Sale Sold out
Taxes included. Shipping calculated at checkout.
View full details

Overview

The NVIDIA® Jetson AGX Thor™ Developer Kit is a Developer Kit tailored for next‑generation humanoid robotics. It delivers up to 2070 FP4 TFLOPS AI performance with a Blackwell architecture GPU and 128 GB high‑speed memory, all in a compact 130 W power envelope. Compared to Jetson AGX Orin™, it offers up to 7.5× more AI performance and 3.5× better energy efficiency to run the latest generative AI models in real time for multimodal perception, decision‑making, and control.

Key Features

Extreme AI Performance for Robotics

Powered by the Jetson T5000 module with a Blackwell GPU featuring 2560 cores and 96 Tensor Cores, delivering up to 2070 FP4 / 1035 FP8 TFLOPS—ideal for running a wide range of generative AI models such as LLM, VLA (like NVIDIA® Isaac™ GR00TN), ViT, etc.

Optimized for Transformer and Multimodal Workloads

Built‑in Transformer engine, Multi‑Instance GPU (MIG) support, and 128 GB LPDDR5X memory (273 GB/s) enable efficient training and inference at the edge.

Real‑Time Processing and Control

14‑core Arm® Neoverse®‑V3AE 64‑bit CPU provides fast, deterministic performance for real‑time robotic applications.

High‑Speed Sensor and Network Integration

Supports up to 4× 25 GbE connections, a 140 W power supply, and rich I/O including HDMI, DisplayPort, USB 3.2/USB‑C, CAN, and GbE for high‑bandwidth, low‑latency sensor fusion.

Specifications

AI Performance 2070 TFLOPS (FP4—Sparse)
GPU 2560‑core NVIDIA Blackwell architecture GPU with 96 fifth‑gen Tensor Cores; Multi‑Instance GPU (MIG) with 10 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
CPU Max Frequency 2.6 GHz
Vision Accelerator 1 个 PVA v3
Memory 128 GB 256‑bit LPDDR5X; 273 GB/s
Storage 1 TB NVMe M.2 Key M Slot; Supports NVMe through PCIe; Supports SSD through USB 3.2
Video Encode 6× 4Kp60 (H.265); 12× 4Kp30 (H.265); 24× 1080p60 (H.265); 50× 1080p30 (H.265); 48× 1080p30 (H.264); 6× 4Kp60 (H.264)
Video Decode 4× 8Kp30 (H.265); 10× 4Kp60 (H.265); 22× 4Kp30 (H.265); 46× 1080p60 (H.265); 92× 1080p30 (H.265); 82× 1080p30 (H.264); 4× 4Kp60 (H.264)
Camera HSB camera via QSFP slot; USB camera; Up to 20 cameras via HSB; Up to 6 cameras through 16× 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 Gen5 (x8 lanes); Root port only—C1 (x1) and C3 (x2); Root Point or Endpoint—C2 (x1), C4 (x8), and C5 (x4)
USB* 2× USB‑A (3.2 Gen2); 2× USB‑C (3.1); xHCI host controller with integrated PHY (up to 3× USB 3.2 and 4× USB 2.0)
Networking* 1× 5GBe RJ45 connector; 1× QSFP28 (4× 25 GbE); 4× 25 GbE
Display 1× HDMI 2.0b; 1× DisplayPort 1.4a; 4× 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 lane, I2C, PCIe x2 lane; 2× 13‑pin CAN header; 2× 6‑pin automation header; LED; JTAG connector (2× 5‑pin header); 1× fan connector—12 V, PWM, Tach; Audio panel header (2× 5‑pin); Microfit power jack; RTC backup battery connector 2‑pin; 5× I2S/2× audio hub (AHUB), 2× DMIS, 4× UART, 4× CAN, 3× SPI, 13× I2C, 6× PWM outputs
Power 40 W – 130 W
Mechanical Developer kit: 243.19 mm × 112.40 mm × 56.88 mm; Thermal Transfer Plate (TTP) and optional fan or heat sink. Module: 100 mm × 87 mm; 699‑pin B2B connector; Integrated TTP with heatpipe.

* Refer to the Software Features section of the latest NVIDIA Jetson Linux Developer Guide for supported features.
** Low‑speed I/O spec is subject to change.

What’s Included

  • NVIDIA® Jetson T5000 module with heatsink and reference carrier board ×1
  • DC power 140 W ×1
  • WiFi Module ×1
  • 1 TB NVMe ×1
  • Quick Start Guide ×1

Applications

  • Humanoid robotics development and workflows
  • Spatial intelligence: visual AI agents using NVIDIA Video Summarization and Search (VSS)
  • Multi‑sensor processing and real‑time sensor fusion with NVIDIA Holoscan
  • Generative AI at the edge: multimodality and foundation models

Documents

ECCN/HTS

HSCODE 8543709990
USHSCODE 8517620090
EUHSCODE 8543709099
COO CHINA

Details

Jetson AGX Thor Developer Kit, NVIDIA Jetson T5000 offers 2070 TFLOPS AI, 273 GB/s bandwidth, 14-core CPU, 1TB SSD, Wi-Fi 6e, rich I/O, and advanced cooling for humanoid robots.

NVIDIA Jetson T5000 with Blackwell GPU delivers 2070 TFLOPS AI performance, 273 GB/s memory bandwidth, and 14-core CPU. Features include 1TB NVMe SSD, Wi-Fi 6e, rich I/O connectivity, heatsink, fan, and advanced cooling for humanoid robots.

Jetson AGX Thor Developer Kit, NVIDIA GROOT accelerates humanoid robot development using advanced workflows, demonstrated by two robots in a lab environment.

Humanoid Robotics. Supercharge humanoid robot development with NVIDIA GROOT workflows. Two robots in lab setting.

Jetson AGX Thor Developer Kit, AI agent uses NVIDIA tech for video analysis, tracing 10 years of AI evolution from perception to physical AI, with audio transcription and search capabilities.

Visual AI agent powered by NVIDIA Video Summarization and Search. Traces AI evolution—Perception to Physical AI—over ten years. Includes audio transcription.

Jetson AGX Thor Developer Kit, NVIDIA Holoscan enables real-time, streamlined multi-sensor processing for efficient data integration and analysis.

Multi-Sensor Processing: Streamline sensor processing in real time with NVIDIA Holoscan.

Jetson AGX Thor Developer Kit, Generative AI powers physical world via edge deployment, multimodality, and foundation models.

Generative AI powers physical world via edge deployment, multimodality, and foundation models.