Imagine a tiny computer that can run AI models, control machines, and survive harsh factories-all in a box the size of a credit card. The Firefly AIBOX-OrinNano Industrial Embedded PC, 8GB/128GB Standard without Wifi gives you that power without the clutter of a full PC. Engineers often struggle with slow, bulky hardware that cannot handle modern deep-learning tasks. This device solves that problem by delivering high AI performance in a rugged, compact package.
Firefly AIBOX-OrinNano Industrial Embedded PC, 8GB/128GB Standard without Wifi
Many industrial projects need real-time AI but the available computers are either too big, draw too much power, or cannot survive temperature swings. When a controller fails, production stops and costs rise. The Firefly AIBOX-OrinNano removes those barriers. Its hexa-core ARM Cortex-A78AE CPU and 67 TOPS AI engine run deep-learning models locally, while the 12 V power draw stays under 8 W in normal use. One Gigabit Ethernet port and USB 3.0 keep connectivity simple, and the fan-less design fits tight enclosures.
Key Features
- Hexa-core 64-bit ARM Cortex-A78AE processor up to 1.7 GHz.
- 8 GB DDR4 RAM and 128 GB eMMC storage for fast data handling.
- 67 TOPS AI accelerator supporting CNN, RNN, LSTM and major deep-learning frameworks.
- HDMI 2.1 output (4K @ 60 fps) for high-resolution display needs.
- 1 Gigabit Ethernet (RJ45) for reliable wired networking.
- 2 USB 3.0 ports (1 A max) for fast peripheral connections.
- Type-C USB 2.0 OTG, console serial port, recovery interface, and power button.
- External watchdog support for added system safety.
- Ubuntu 22.04 based Jetson system with CUDA, TensorRT and cuDNN acceleration.
Specifications
| AI Performance |
67 TOPS |
| Traditional Network Architectures |
Supports traditional architectures such as CNN, RNN, and LSTM; supports deep learning frameworks including TensorFlow,
PyTorch, MXNet, PaddlePaddle, ONNX, and Darknet; supports custom operator development
|
| Container Management |
Docker container management technology facilitates easy image deployment |
| Large Model Support |
Robotic Models: ROS models
Large Language Models: Private deployment of ultra-large-scale Transformer-based 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
|
| CPU |
Hexa-core 64-bit ARM Cortex-A78AE v8.2 processor, up to 1.7GHz |
| Power Input |
DC 12V (DC 5.5 × 2.1mm) |
| Power Consumption |
Normal: 7.2W (12V/600mA)Maximum: 18W (12V/1500mA) |
| Video Output |
1 × HDMI 2.1 (4K @ 60fps) |
| Operating System |
Jetson system based on Ubuntu 22.04 with a comprehensive desktop Linux environment, accelerated graphics, and
support for NVIDIA CUDA, TensorRT, and cuDNN libraries
|
| Dimensions |
93.4 mm × 93.4 mm × 50 mm |
| Weight |
≈ 500g |
| Environment |
Operating Temperature: -20℃ to 60℃
Storage Temperature: -20℃ to 70℃
Storage Humidity: 10% to 90% RH (non-condensing)
|
| Watchdog |
Supports external watchdog |
| Ethernet |
1 × Gigabit Ethernet (1000Mbps / RJ45) |
| USB |
2 × USB 3.0 (Maximum: 1A) |
| Other Interfaces |
1 × Type-C (USB 2.0 OTG)
1 × Console (debug serial port)
1 × Recovery interface
1 × Power button
|
