Precision You Can Trust
NABL-Accredited Calibration | Available on Request
The Hailo-8R mPCIe AI Acceleration Module (HMP1RB1C2XAE) delivers up to 13 TOPS of on-device AI inference performance in a standard PCI Express Mini Card Full-Mini form factor. Using a PCIe Gen 3.0 x1 interface, the extended-temperature AI accelerator supports real-time computer vision, video analytics, machine vision, industrial inspection, robotics and other edge AI applications..
The Hailo-8R mPCIe AI Acceleration Module HMP1RB1C2XAE is a compact edge AI accelerator starter kit designed to add dedicated neural-network inference capability to embedded computers, industrial PCs, edge gateways and other systems equipped with an mPCIe Full-Mini socket.
Powered by the Hailo-8R AI processor, the module provides up to 13 Tera Operations Per Second (TOPS) of AI compute performance. This enables deep-learning workloads to execute directly on the edge, helping reduce latency, cloud dependency and network bandwidth requirements while supporting real-time decision-making.
The HMP1RB1C2XAE uses the standard PCI Express Mini Card Full-Mini F1 form factor with dimensions of 30 × 50.95 mm. This standardised module format allows developers and system integrators to add AI acceleration to compatible systems without requiring a custom accelerator board design.
Host communication is provided through a PCIe Gen 3.0 x1 interface, enabling high-speed data transfer between the Hailo-8R accelerator and the host processor. The module supports both x86 and ARM-based host architectures, making it suitable for a wide variety of embedded and edge computing platforms.
The Hailo-8R is designed for real-time, low-latency AI inference and supports simultaneous multi-stream and multi-model processing. This makes it suitable for applications that need to analyse multiple video streams or execute several neural networks concurrently, including object detection, classification, industrial visual inspection, intelligent video analytics and smart-camera applications.
Hailo provides a comprehensive software ecosystem for model development and deployment. The module supports popular AI frameworks including TensorFlow, TensorFlow Lite, Keras, PyTorch and ONNX, together with Hailo's Dataflow Compiler, HailoRT runtime and Model Zoo resources.
The extended-temperature Hailo-8R mPCIe platform supports an operating range from -40°C to +85°C, making it suitable for industrial automation, transportation, outdoor edge systems, rugged embedded platforms and other demanding environments.
The HMP1RB1C2XAE starter kit configuration includes the Hailo-8R mPCIe module together with thermal-management components and installation accessories, providing developers with the hardware required to evaluate and integrate Hailo AI acceleration into compatible systems.
Up to 13 TOPS AI inference performance
Manufacturer Part Number: HMP1RB1C2XAE
Powered by the Hailo-8R AI processor
PCI Express Mini Card Full-Mini F1 form factor
Compact 30 × 50.95 mm dimensions
PCIe Gen 3.0 x1 host interface
Designed for real-time, low-latency edge AI inference
Supports simultaneous multi-stream processing
Supports multi-model neural-network inference
Compatible with x86 and ARM host platforms
Supports Linux and Windows
Compatible with TensorFlow, TensorFlow Lite, Keras, PyTorch and ONNX
Extended operating temperature from -40°C to +85°C
Suitable for industrial and embedded edge systems
Designed for efficient AI processing at the edge
Standard mPCIe form factor simplifies hardware integration
CE and FCC certified
Starter kit includes thermal-management components and installation accessories
| Specification | Details |
|---|---|
| Manufacturer | Hailo Technologies |
| MPN | HMP1RB1C2XAE |
| Processor | Hailo-8R |
| AI Performance | Up to 13 TOPS |
| Module Type | mPCIe AI Acceleration Module / Starter Kit |
| Form Factor | PCI Express Mini Card Full-Mini F1 |
| Dimensions | 30 × 50.95 mm |
| Host Interface | PCIe Gen 3.0 x1 |
| Host Architecture Support | x86, ARM |
| Operating Systems | Linux, Windows |
| AI Frameworks | TensorFlow, TensorFlow Lite, Keras, PyTorch, ONNX |
| Operating Temperature Min. | -40°C |
| Operating Temperature Max. | +85°C |
| Certifications | CE, FCC |
| Multi-Stream Inference | Supported |
| Multi-Model Inference | Supported |
| Primary Workload | Deep Learning / Neural Network Inference |
| Deployment Type | Edge / Embedded AI |