Imagine you need AI power at the edge but you cannot fit a noisy, bulky box into your project. The Firefly EC-OrinNX Jetson Orin NX Edge AI Fanless PC, 16GB/128GB Without wifi gives you a compact, silent computer that runs deep-learning models right where the data is created. It solves the common problem of overheating and space limits in industrial or field deployments, letting you run vision, language, or robotics AI without a cooling fan.
Firefly EC-OrinNX Jetson Orin NX Edge AI Fanless PC, 16GB/128GB Without wifi
Many edge devices struggle with heat, power spikes, and the need for a full desktop PC. When a device overheats, it shuts down and your AI task stops, causing costly downtime. This problem matters because every second of missed inference can mean lost data or delayed decisions. The Firefly PC removes the fan, uses a low-voltage 12 V supply, and packs an octa-core CPU with a 1024-core GPU, delivering 157 TOPS in a box the size of a small router. It gives you reliable performance, quiet operation, and the ability to mount it in tight enclosures.
Key Features
- Octa-core Arm Cortex-A78AE CPU up to 2.0 GHz.
- 1024-core NVIDIA Ampere GPU with 32 Tensor Cores.
- 157 TOPS AI performance for heavy inference workloads.
- 16 GB DDR4 RAM and 128 GB eMMC storage, expandable via M.2 NVMe SSD.
- DC 12 V power input, compatible with 9-24 V sources.
- Audio jack with mic support, suitable for voice-activated applications.
- SIM slot and Phoenix connector offering RS485, RS232, CAN 2.0 interfaces.
- Operating temperature from -20 C to 60 C and humidity up to 90 % RH.
Specifications
| AI Performance |
157 TOPS |
| CPU |
Octa-core Arm® Cortex®-A78AE v8.2 64-bit CPU, up to 2.0GHz |
| GPU |
1024-core NVIDIA Ampere architecture GPU with 32 Tensor Cores |
| Storage |
1 × M.2 (internal device, PCIe NVMe SSD expansion available, supporting 2242/2260/2280) |
| Power |
DC 12V (5.5mm × 2.1mm, 9-24V wide input voltage) |
| Audio |
1 × 3.5mm audio jack, supporting MIC recording, CTIA standard |
| OS |
Based on Ubuntu 22.04, the Jetson system provides a comprehensive desktop Linux environment with accelerated graphics,
supporting NVIDIA CUDA 11.4.19, TensorRT 8.5.2, and CuDNN 8.6.0
|
| Software Support |
Robotic Models: ROS models
Large Language Models: Private deployment of Transformer-based ultra-large-scale parameter models such as LLaMA2, ChatGLM, and Qwen
Vision Models: ViT, Grounding DINO, and SAM
AI Painting: Stable Diffusion V1.5 image generation model in the AIGC field
Deep Learning Frameworks: TensorFlow, PyTorch, MXNet, ONNX, PaddlePaddle, Darknet, and custom operator development
Traditional Network Architectures: CNN, RNN, and LSTM
Docker Containerization: Docker container management technology for easy image deployment
|
| Dimension |
188mm × 88.44mm × 50.65mm |
| Environment |
Operating Temperature: -20℃～60℃
Operating Humidity: 10%～90%RH (non-condensing)
|
| Network |
Ethernet: 1 × Gigabit Ethernet (RJ45)
|
| USB |
2 × USB 3.0 |
| Other |
1 × Type-C (USB 2.0/DEBUG)
1 × SIM Card1 × Phoenix connector (2 × 4Pin, 3.5mm pitch): 1 × RS485, 1 × RS232, 1 × CAN 2.0
|
