Geniatech AIM-H8 is a compact, power-efficient AI acceleration module based on the Hailo-8 NPU, delivering up to 26 TOPS with a TDP of just 2.5W. It features a Board-to-Board (B2B) connector for flexible integration into custom carrier boards. With ultra-low latency and support for multi-stream inferencing, it enables real-time performance for edge AI computing applications while minimizing power consumption.
| Chip | Hailo-8 |
| AI performance | 26 TOPS(INT8) |
| Host Processor | NXP,Renesas,TI,Rockchip,SocioNext,Xilinx |
| AI Model Frameworks Supported | TensorFlow,TensorFlow Lite,Keras,PyTorch&ONNX |
| Memory | Processor integration |
| Host Interface | B2B(4-lane PCIE Gen3) |
| OS Support | Linux, Windows |
| Power Consumption(Typical) | 5W |
| Operating Temperature | -40℃~85℃ |
| Dimensions | 40×55mm |