NVIDIA Jetson Orin Nano Developer Kit with Compute Carrier & MIPI CSI-2 Sensors for AI Robotics

Next-Level Generative AI at the Edge

The NVIDIA Jetson Orin Nano Super Developer Kit brings generative AI and multimodal workloads to the edge in a compact, developer-friendly form. Delivering up to 67 TOPS of AI performance with 102 GB/s memory bandwidth, it’s capable of running vision transformers, language models, and advanced sensor fusion in real time. It delivers a major leap in performance—1.7X faster than the original Orin Nano Developer Kit—and is powered by the NVIDIA AI software stack and ecosystem.

Built for Experimentation and Scale

At its core, the kit includes a Jetson Orin Nano 8GB module mounted on an AI & Robotics Reference Carrier Board. The module features an Ampere GPU with 1024 CUDA cores and 32 Tensor Cores, a 6-core Arm Cortex-A78AE CPU, 8GB of LPDDR5 memory, and microSD storage support. With a configurable 7W–25W power envelope, developers can tune performance or efficiency depending on the application. This hardware combination makes it possible to prototype devices that demand multiple concurrent AI pipelines, from real-time vision to conversational AI.

Expandable I/O for Real Projects

The included AI & Robotics Reference Carrier Board offers broad connectivity for peripherals, expansion cards, and sensors—designed specifically for edge AI development. With dual MIPI CSI-2 camera connectors, dual M.2 slots, Gigabit Ethernet, USB 3.2 Gen2, DisplayPort, and a 40-pin expansion header, it provides the flexibility needed for robotics, drones, smart vision systems, and other AI-powered devices.

Backed by the NVIDIA AI Software Stack

Every Orin Nano module runs NVIDIA’s AI software ecosystem, including Isaac for robotics, Metropolis for vision AI, and Holoscan for real-time sensor processing. Tools like Omniverse Replicator for synthetic data generation and the TAO Toolkit for fine-tuning pretrained models streamline the journey from idea to deployment.

• 67 TOPS AI performance for generative models, transformers, and multimodal pipelines.
• NVIDIA Ampere GPU with 1024 CUDA cores and 32 Tensor Cores.
• 6-core Arm Cortex-A78AE CPU for multi-threaded processing.
• 8GB LPDDR5 memory, 128-bit, 102 GB/s bandwidth.
• microSD slot plus NVMe expansion via M.2 Key M.
• Camera input: dual MIPI CSI-2 connectors for high-resolution sensors.
• Video encode: up to 4K30 / multiple 1080p60 streams.
• Video decode: up to 8K30 / 4K60 / 1080p high-stream density.
• Display output: DisplayPort 1.4 (carrier board).
• Connectivity: 4× USB 3.2 Gen2, USB-C, Gigabit Ethernet, Wi-Fi module (included).
• Power modes configurable between 7W and 25W.
• Developer kit includes module with heatsink, reference carrier board, PSU, Wi-Fi, and quick start guide.
• Compatible with Orin Nano and Orin NX modules for scalable prototyping to production.
Carrier Board Specifications