Prev
Next
AIB-AT78/68
• Powered by NVIDIA® Jetson T5000™ / Jetson T4000™
• NVIDIA Blackwell GPU & 14/12 -Cores ARM Neoverse V3AE CPU
• AI Power up to 2070/1200 FP4 TFLOPS
• Support QSFP28 up to 100Gbps
• Support Dual 10G LAN port
• Support EtherCAT
• 4 x USB 3.2 Gen2
• Wide range DC-in voltage
 |
NVIDIA Jetson Thor™ |
Overview
Unlock Edge Supercomputing for Next-Gen Robots with Aetina AIB-AT Series
The Aetina AIB-AT Series is a compact edge supercomputing platform powered by NVIDIA® Jetson Thor™, delivering ultimate compute performance for next generation general purpose robots. Designed for real-time sensor fusion, multi-model synergy and ultra-low-latency motion control, the AIB-AT Series empowers robots to perceive, reason and act with unprecedented speed and intelligence.
The series includes two autonomous edge AI platforms:
- AIB-AT78, powered by NVIDIA Jetson T5000™ module, delivers up to 2070 FP4 TFLOPS
- AIB-AT68, powered by NVIDIA Jetson T4000™ module, delivers up to 1200 FP4 TFLOPS
AIB-AT78/68 Product Benefits
| Product Benefits | Description |
|---|---|
| Robotic Brain | Aetina’s supercomputing platforms serve as the central intelligence for general-purpose robots, accelerating innovation and large-scale deployment of next-generation robotic systems. |
| Supercomputing for LLM Multi-Model Synergy | Powered by NVIDIA Jetson T5000 / Jetson T4000 with up to 2070 FP4 TFLOPS / 1200 FP4 TFLOPS and 128 GB LPDDR5X high-speed memory, the platform enables ultra-low-latency inference of large language models (LLMs) on robots—supporting real-time perception, reasoning, and action. |
| High-Speed Motion Control Connectivity | Supports dual RJ45 10GbE ports and one RJ45 1GbE EtherCAT port, providing flexible connectivity for motion control systems |
| Advanced Sensor Fusion | • Support one super-fast QSFP28 connectivity and NVIDIA Holoscan bridge for seamless integration of cameras and diverse robotic sensors. • Support up to 8x GMSL cameras via 120-pin MIPI interfaces, enabling multi-view perception |
| Networking and Positioning | • Support M.2 B key for 4G, 5G and GPS with 1PPS for precise positioning, synchronization, and route planning in autonomous systems |
| Edge Ready Design | • Compact form factor：L:165 mm x W:145.53mm x H:42.78mm (w/o fantsink) • Wide voltage input：9–48V • Wide operating temperature：–25°C to 70°C, ensuring reliable operation in demanding environments |
| Full NVIDIA Software Stack | Support NVIDIA Jetpack™ 7.0, Isaac™ AI robot development platform, NVIDIA Isaac GR00T foundation model for generalized humanoid robot reasoning, NVIDIA Metropolis vision AI platform, providing a comprehensive software foundation for robotic innovation. |
| Targeted Applications | • Humanoid robots for smart manufacturing and service applications • Healthcare and assistive robots • Precision agriculture robots • Logistics robots and AMRs • Autonomous underwater vehicles (UUVs) |
Specifications
| Model Number | AIB-AT78 | AIB-AT68 |
| Module Compatibility |
NVIDIA Jetson Thor T5000™ |
NVIDIA Jetson Thor T4000™ |
| AI Performance |
2070 FP4 TFLOPS / 1035 FP8 TFLOPS |
1200 FP4 TFLOPS |
| GPU | 2560-core NVIDIA Blackwell architecture GPU with 96 fifth-gen Tensor Cores |
1536-core NVIDIA Blackwell architecture GPU with 64 fifth-gen Tensor Cores |
| CPU |
14-Core ARM Neoverse V3AE (64-bit) |
12-Core ARM Neoverse V3AE (64-bit) |
| Memory |
128GB 256-bit LPDDR5X |
64GB 256-bit LPDDR5X |
| Storage | 1 x M.2 M-Key 2280 (NVMe 1TB built-in) | |
| Display | 1 x HDMI 2.0b 1 x DP 1.4a |
|
| LAN | 2 x RJ45 10GbE 1 x RJ45 1GbE EtherCAT |
|
| USB | 1 x USB-C for OTG only 4 x USB-A 3.2 Gen 2 |
|
| I/O Interfaces | 1× QSFP28 slot (T5000: 4× 25 GbE, T4000: 3× 25 GbE) 1 x HD Audio-in/out 2 x RS232/RS485/RS422 4 x CAN FD (Isolation, Support for T5000 only) 4 x DI/DO (Isolation) @3.3V 2 x UART @ 3.3V 2 x I2C @ 3.3V 1 x OOB (by request) |
|
| Expansion | 1 x M.2 M-Key 2280 (NVMe1TB built-in) 1 x M.2 E-Key 2230 (WiFi/BT) 1 x M.2 B-Key 3042/3052 (LTE/4G/5G) 1 x Micro SIM card slot |
|
| Camera | 1 x 120pin MIPI | |
| Power Input/Connector | DC-in 9V-48V | |
| Dimension (WxDxH) | L:187.4 mm x W:145.53mm x H:42.78mm (w/o fantsink, with bracket) L:213.7 mm x W:145.53mm x H:83.44mm (with fantsink & bracket) |
|
| Net Weight | TBD kg | |
| Temperature | Operating Temp.: -25°C ~ +70°C (-13°F ~ +158°F) Storage Temp.: -40°C ~ +85°C (-40°F ~ +185°F) |
|
| Humidity | 95% (non-condensing) | |
| Software Support | Linux | |
| Certification | CE / FCC Class A / UKCA | |
*Product specifications are subject to change without prior notice
Why Aetina
| Value-Added Points | Description |
|---|---|
| NVIDIA Elite Partner | Priority access to NVDIA technologies and Jetson products, helping clients innovate faster and scale within the NVIDIA ecosystem. |
| Customization Services | From chassis and thermal design to BSP integration, Aetina provides tailored solutions to meet every unique deployment requirement. |
| Developer Kit | The AIB-THOR-ES Preview Kit, powered by NVIDIA Jetson T5000, enables AI developers to accelerate POC validations and project development. |
| Collaborative AI Deployments | With over a decade of edge AI expertise and a robust ecosystem, Aetina provides comprehensive technical and team support to simplify integration and accelerate time to deployment.
|
| Rich Set of Validated Peripherals |
Backed by NVIDIA ecosystem partners and the Innodisk Group, Aetina offers a wide range of fully validated peripherals—including storage, cameras, Wi-Fi, and 4G/5G—reducing integration time and overall system cost. |
FAQ
| Question | Answer |
|---|---|
| What is NVIDIA Jetson Thor? | NVIDIA Jetson Thor is NVIDIA’s next-generation edge AI computing platform designed for advanced robotics. It delivers supercomputer-class performance at the edge, enabling real-time multimodal AI, large-scale model inference, and deterministic control for autonomous systems. |
| What is the difference between NVIDIA Jetson Thor and Orin? | NVIDIA Jetson Thor represents a significant leap beyond Jetson Orin™, offering substantially higher AI compute performance—up to 7.5X the AI performance of Jetson AGX Orin™, along with enhanced support for generative AI workloads and improved capabilities for complex robotics tasks such as real-time LLM inference, sensor fusion, and ultra-low-latency control. |
| Where can I purchase an NVIDIA Jetson Thor computing platform? | • For standard products, the AIB-AT series is scheduled to be available by the end of Q1 2026. • For development kits, AIB-THOR-ES preview kits are available now! (Watch Unboxing Video) Please contact Aetina for more information. |
Specifications
Dimensions
Ordering Information
| Model Number | Description |
|---|---|
|
AIB-AT78-1-A1 |
THOR T5000, 1 x PD, 1 x HDMI, 1 x GbE, 2 x 10GbE, 5 x USB, QSFP, B/E/M Key, 9 to 48 VDC input, -25°C ~ +70°C |
|
AIB-AT68-1-A1 |
THOR T4000, 1 x PD, 1 x HDMI, 1 x GbE, 2 x 10GbE, 5 x USB, QSFP, B/E/M Key, 9 to 48 VDC input, -25°C ~ +70°C |
Downloads
| Model Name | File Name | OS | Version | Update Date | Download |
|---|---|---|---|---|---|
| AIB-AT78/68 | Aetina_AIB-AT78/68_user manual | v1.0 | 2026-07-09 | Download | |
| AIB-AT78/68 | Aetina_AIB-AT78/68_Jetson Thor_datasheet | v1.1 | 2026-07-09 | Download | |
| ARM-based edge AI systems | Aetina_DeviceEdge_brochure_EN_2026 | v1.0 | 2026-02-26 | Download |