MPN: EC-OrinNX
Firefly EC-OrinNX
157 TOPS · Orin NX · Edge AI · Fanless

NVIDIA Jetson Orin NX 16GB · 128GB NVMe SSD · 8-Core ARM · 1024 CUDA Cores · GbE · WiFi 6 · USB 3.2 · CAN · Wide Temp

Specifications

Module

NVIDIA Jetson Orin NX
16GB Version

AI Performance

Up to 157 TOPS
(100 TOPS Standard)

GPU

1024 CUDA Cores
32 Tensor Cores
NVIDIA Ampere Architecture

CPU

8-core Arm Cortex-A78AE
16GB LPDDR5 Memory

Storage

128GB NVMe SSD
(M.2 2280 slot)

Networking

GbE RJ45 LAN
WiFi 6 · BT 5.2

Display

HDMI 2.1 Output
Up to 8K60

Video Decode

H.265: 1×8K30, 2×4K60
4×4K30, 9×1080p60
18×1080p30

I/O

USB 3.2 · USB-C
RS-232/485 · CAN
GPIO Header

Expansion

M.2 2280 NVMe
M.2 2230 WiFi/BT
MicroSD Slot

Enclosure

Aluminum · Fanless
Compact Form Factor

Power

Wide DC Input
9-36V Range

Operating Temp

Industrial Wide Temp
-20°C ~ 60°C

AI Frameworks

TensorFlow · PyTorch
OpenCV · JetPack SDK
ROS · Docker

Generative AI

LLM · Vision Models
Stable Diffusion
Transformer Models

OS Support

NVIDIA JetPack 5.x+
Ubuntu 20.04/22.04

Overview

The Firefly EC-OrinNX is a compact yet powerful edge AI fanless PC built on the NVIDIA Jetson Orin NX core module (16GB variant). Designed by Firefly — a Shenzhen-based embedded computing specialist known for the ROC-RK3588 and AIO-3588 series — the EC-OrinNX delivers up to 157 TOPS of AI acceleration, making it suitable for demanding edge inference workloads from object detection and tracking to generative AI models including LLMs and vision transformers.

At its heart, the Jetson Orin NX module integrates an 8-core Arm Cortex-A78AE CPU, a 1024-core NVIDIA Ampere GPU with 32 Tensor cores, and 16GB of high-bandwidth LPDDR5 memory. The EC-OrinNX pairs this with 128GB of fast NVMe SSD storage via an M.2 2280 slot, along with a MicroSD slot for secondary storage. Network connectivity includes Gigabit Ethernet, WiFi 6, and Bluetooth 5.2 for flexible deployment in factory, warehouse, and smart city environments.

With HDMI 2.1 output supporting up to 8K60, the EC-OrinNX is well-suited for high-resolution video analytics and digital signage applications. Its H.265 hardware decoder handles up to 18 simultaneous 1080p30 streams, making it ideal for multi-camera video surveillance and AI-driven video processing. Rich I/O includes USB 3.2, USB-C, RS-232/485 serial, CAN bus, and GPIO headers for industrial sensor and actuator integration. The fanless aluminum enclosure, wide 9-36V DC input, and -20°C to 60°C operating range ensure reliable 24/7 operation in industrial environments. The EC-OrinNX supports the full NVIDIA JetPack SDK ecosystem, running TensorFlow, PyTorch, OpenCV, ROS, and Docker containerized workloads at the edge.

Need Firefly EC-OrinNX?

Contact QS Compute for pricing and availability. Global B2B supply.

Request Quote