Fogwise® AirBox
Easy to Build Your Own GPT and Stable Diffusion
Powered by SOPHON SG2300x
- Octa-core A53,
up to 2.3GHz - Support 12 channels HD
hardware encoding - Support 32 channels HD
hardware decoding - Support INT8,
FP16/BF16 and FP32 - Mainstream frameworks Support :
- Built-in 32TOPS@INT8 high-performance TPU (Tensor Processing Unit)
Capable of Operating as An Embedded Device in
Harsh Environments
Aluminum alloy shell, Protects equipment in harsh environments
Rugged
and Drop-proof
Corrosion
Resistant
High Thermal
Conductivity
Active heat dissipation
PWM speed control
Aluminum Alloy Case: Durable and Reliable
Easy to assemble
Easy-to-Use, Convenient and Efficient
BMNNSDK one-stop toolkit provides a series of software tools including the underlying driver
environment, compiler and inference deployment tool.
The easy-to-use and convenient toolkit covers the model optimization, efficient runtime support and
other capabilities required for neural network inference.
Support 32 Channels HD Hardware Decoding
Support 32-channel 1080P@25fps HD video hard decoding
Supports Multiple Storage Devices
64GB eMMC Onboard
MicroSD Slot
M.2 M Key for NVMe SSD
Radxa Model Zoo
Applications
Public Security
Private GPT
Private Stable Diffusion
Smart Healthcare
Industrial AI
Fogwise® AirBox Works with
Radxa Power PD 65WCharge Smarter, Faster