NVIDIA Jetson Orin Nano Super
Expert rating9 / 10
- AI Performance:
- Up to 67 INT8 TOPS
- Power Modes:
- 7W / 15W Standard
- Software:
- NVIDIA JetPack SDK (TensorRT
- Camera Support:
- Dual MIPI CSI Connectors
Consumercarts is reader-supported. Learn more
Disclaimer: The price and specification of the given product may differ from the actual product. Please confirm on the retailer's website before buying.
Manufacturer: NVIDIA Last updated: September 9, 2026
NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support) Review
The NVIDIA Jetson Orin Nano Super Developer Kit delivers up to 67 INT8 TOPS of AI performance, runs at 7W and 15W standard power modes with a MAXN mode reaching up to 25W, ships with the NVIDIA JetPack SDK covering TensorRT, cuDNN and CUDA, and provides dual MIPI CSI camera connectors in a compact developer-kit form factor.
It is independently confirmed as one of the top options for AI and machine learning projects, handling AI inference, computer vision and robotics workloads considerably better than a standard Raspberry Pi. Current guidance is direct on this point: only the NVIDIA Jetson line is a genuinely real AI board among current SBC options.
Processor and Performance
The "Super" update specifically unlocks a MAXN power mode reaching up to 25W, up from the standard 7W and 15W envelope modes, delivering the 67 INT8 TOPS figure that puts this board's AI inference capability well beyond what ARM-based general-purpose boards can approach.
Memory and Storage
Sustained AI inference workloads at the higher power modes generate real heat, and community feedback consistently flags active cooling as necessary for sustained performance at the MAXN power tier, a genuine practical consideration when planning a build around this board.
Connectivity and I/O
Dual MIPI CSI connectors support high-resolution camera modules directly, a meaningful advantage for computer vision and robotics projects that need native camera input without USB capture card workarounds.
Software and Ecosystem Support
NVIDIA's JetPack SDK provides pre-optimized libraries for TensorRT, cuDNN, and popular frameworks including PyTorch and TensorFlow, with full CUDA support giving this board access to the same deep learning software ecosystem used on NVIDIA's much larger GPU platforms, a genuine, substantial advantage over ARM boards with only limited NPU software support.
Comparison to Sibling Products
Against the Orange Pi 5 Plus and Radxa ROCK 5B+, both covered elsewhere on this site, those boards include a 6 TOPS NPU that works for specific use cases like Frigate object detection, but current guidance is direct that the software story is rough; for real AI work, the Orin Nano Super remains the recommended choice. Against the Raspberry Pi 5, also covered elsewhere on this site, that board offers unmatched general-purpose community support but no comparable AI inference capability at all.
Specifications
Identity
| Brand | NVIDIA |
|---|
Graphics and AI
| AI Performance | 67 TOPS |
|---|
| Model | NVIDIA Jetson Orin Nano Super Developer Kit |
|---|---|
| Power Modes | 7W / 15W standard, up to 25W in MAXN mode |
| Software | NVIDIA JetPack SDK (TensorRT, cuDNN, CUDA) |
| Camera Support | Dual MIPI CSI connectors |
Expert Review
Reviewed by Burooj Alam. Rated by us, not by users.
Expert rating9 / 10
The only single-board computer in this class capable of genuinely serious AI inference, provided you budget for the price, power and active cooling.
Best for
- AI researchers, robotics engineers, and computer vision developers — the only board in this class capable of genuinely serious AI inference work.
- Projects requiring native high-resolution camera input — the dual MIPI CSI connectors are a real, specific advantage.
Not for
- Simple projects or buyers on a tight budget — the higher price, power requirements and active cooling needed at sustained MAXN performance make it a poor fit; the Raspberry Pi 5 is the better choice.
FAQs
What are the key specifications of the NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)?
The NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support) is a single board computer from NVIDIA. Key specs: AI Performance 67, Power Modes 7W / 15W standard, up to 25W in MAXN mode, Software NVIDIA JetPack SDK (TensorRT, cuDNN, CUDA), Camera Support Dual MIPI CSI connectors.
Who is the NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support) best suited to?
It suits ai researchers, robotics engineers, and computer vision developers — the only board in this class capable of genuinely serious ai inference work., projects requiring native high-resolution camera input — the dual mipi csi connectors are a real, specific advantage..
What are the drawbacks of the NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)?
It is a poor fit for simple projects or buyers on a tight budget — the higher price, power requirements and active cooling needed at sustained maxn performance make it a poor fit; the raspberry pi 5 is the better choice..
Compare and read more
- FriendlyElec NanoPC-T6 (Single-Board Computer/Rockchip RK3588/Octa-Core/Up to 16GB LPDDR5/Dual 2.5GbE and M.2 NVMe) vs NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)
- FriendlyElec NanoPi R6S (Single-Board Computer/Rockchip RK3588S/Octa-Core/Dual 2.5GbE/6 TOPS NPU Router Board) vs NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)
- Banana Pi BPI-R4 (Single-Board Computer/MediaTek Filogic 880/Quad-Core Cortex-A73/WiFi 7/10GbE Router Board) vs NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)
- ASUS NUC 13 Pro Board (Single-Board Computer/Intel Core i7-1360P (Raptor Lake)/Up to 64GB DDR5/Embedded NUC Motherboard) vs NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)
- Google Coral Dev Board (Single-Board Computer/NXP i.MX 8M/Quad Cortex-A53/Edge TPU AI Accelerator/1GB LPDDR4) vs NVIDIA Jetson Orin Nano Super (Single-Board Computer/AI Development Kit/Up to 67 INT8 TOPS/CUDA and TensorRT Support)