• Black circuit board of nvidia igx t5000 ai module with central green to yellow gradient sensor and metallic black components
1 1

NVIDIA

NVIDIA 900-13834-0000-0A1 | Jetson Thor T4000 AI Module, 64GB, 12-Core, USA Made

900-13834-0000-0A1

Stock Status: Enquire

$6,578.00 inc. GST
$5,980.00 ex. GST
Sale Sold out
Shipping calculated at checkout.
20-min response

Need a quote or bulk pricing? Answered within 20 minutes, Sydney business hours.

Description

The Jetson Thor T4000 is a compact Blackwell-based system-on-module for physical AI, robotics and humanoid applications. It suits developers and OEMs building real-time, multi-sensor edge AI systems that need generative AI performance in a small, power-configurable package.

The module delivers up to 1200 FP4 sparse TFLOPS (600 FP8 TFLOPS) from a 1536-core NVIDIA Blackwell GPU with 64 fifth-generation Tensor Cores and Multi-Instance GPU support. A 12-core Arm Neoverse-V3AE 64-bit CPU runs at up to 2.6 GHz, backed by 64 GB of 256-bit LPDDR5X memory at 273 GB/s. Networking is handled by 3x 25 GbE, and power is configurable between 40 W and 70 W. This SKU is the USA Made variant with 64GB and 12 cores, identified by the -0A1 suffix and distinct from sibling SKUs 900-13834-0000-000 and -001.

It connects to a carrier board through a 699-pin B2B connector and measures 100 mm x 87 mm, with an integrated thermal transfer plate and heatpipe. Software is provided through the NVIDIA Jetson Linux and JetPack SDK, with Holoscan Sensor Bridge support for up to 20 cameras and NVMe storage over PCIe.

Enquire with XS Network Tech for new or used pricing and availability.

Features

- Over 4x the AI compute of Jetson AGX Orin with high energy efficiency
- Blackwell GPU with transformer engine for generative AI at the edge
- Multi-Instance GPU support for partitioning workloads
- Low-latency multi-sensor processing with a 12-core CPU, 3x 25 GbE and extensive I/O
- Supports the NVIDIA Isaac platform and GR00T foundation models for humanoid robotics
- Integrated thermal transfer plate with heatpipe
- Configurable 40 W to 70 W power profiles to balance performance and efficiency

Warranty

All products sold by XS Network Tech include a 12-month warranty on both new and used items. Our in-house technical team thoroughly tests used hardware prior to sale to ensure enterprise-grade reliability.

All technical data should be verified on the manufacturer data sheets.

View full details

specs-tabs

Technical Specifications

FAQs

Technical Specifications

- AI performance: 1200 TFLOPS (FP4 sparse), 600 TFLOPS (FP8)
- GPU: 1536-core NVIDIA Blackwell with 64 fifth-gen Tensor Cores
- Multi-Instance GPU: 6 TPCs
- CPU: 12-core Arm Neoverse-V3AE 64-bit, up to 2.6 GHz
- CPU cache: 64 KB I-Cache + 64 KB D-Cache, 1 MB L2 per core, 16 MB shared L3
- Vision accelerator: 1x PVA v3
- Memory: 64 GB 256-bit LPDDR5X, 273 GB/s
- Storage: NVMe through PCIe, SSD through USB 3.2
- Networking: 3x 25 GbE
- PCIe: Up to 8 lanes Gen5
- USB: 3x USB 3.2, 4x USB 2.0
- Display: 4x shared HDMI 2.1, DisplayPort 1.4a HBR2 with MST
- Camera: C-PHY 2.1 and D-PHY 2.1, up to 20 cameras via HSB, up to 32 using virtual channels
- Video encode: 6x 4Kp60 (H.265), 24x 1080p60 (H.265)
- Video decode: 4x 8Kp30 (H.265), 10x 4Kp60 (H.265)
- Other I/O: 13x I2C, 3x SPI, 4x CAN, 4x UART, 5x I2S, 6x PWM outputs
- Power: 40 W to 70 W configurable
- Dimensions: 100 mm x 87 mm
- Connector: 699-pin B2B with integrated thermal transfer plate and heatpipe
- Security: 128-bit AES memory encryption, TrustZone

FAQs

Q: Does the Jetson Thor T4000 need a separate licence?
A: No separate licence is stated for the module. Software is provided through the NVIDIA Jetson Linux and JetPack SDK.

Q: What power does the module need?
A: Power is configurable between 40 W and 70 W. The full 1200 FP4 TFLOPS is delivered at the 70 W profile.

Q: How does the module connect to a system?
A: It uses a 699-pin B2B connector and mounts to a carrier board. The module measures 100 mm x 87 mm and includes an integrated thermal transfer plate.

Q: What storage and cameras does it support?
A: Storage is supported through NVMe over PCIe or SSD over USB 3.2. Up to 20 cameras are supported via the Holoscan Sensor Bridge.

Q: How does this SKU differ from its sibling SKUs?
A: This is the USA Made variant with 64GB memory and a 12-core CPU, identified by the -0A1 suffix. The sibling SKUs are 900-13834-0000-000 and 900-13834-0000-001.

Related Products

View all Enterprise GPUs & Graphics Cards

Recently Viewed

  • Request a Quote

    Looking for competitive pricing? Submit a request, and our team will provide a tailored quote that fits your needs.

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Call us Now  

  • Contact Us Directly

    Have a question or need immediate assistance? Call us for expert advice and real-time support.

    Contact us