ASUS UGen300 M2 AI Module 2GB LPDDR4X

SKU: UGEN300-2GLD4X-M2

Sale price SGD $209.00
Save 13% Regular price $239.00

The ASUS UGen300 M2 AI Module features a Hailo-10H processor and 2GB LPDDR4X memory for local vision AI and small language model inference. Its M.2 2280 Key-M design provides PCIe 3.0 x4 connectivity, up to 40 TOPS at INT4 precision and 2.5W typical power consumption.

 


ASUS UGen300 M2 AI Module 2GB LPDDR4X

SKU: UGEN300-2GLD4X-M2
  • Hailo-10H AI Acceleration: Designed for local vision AI and generative AI applications, including supported small language models.
  • Up to 40 TOPS: Provides 40 TOPS at INT4 precision and 20 TOPS at INT8 precision for supported inference workloads.
  • 2GB LPDDR4X Memory: Integrated memory operates at 4266 MT/s for onboard AI processing.
  • Compact M.2 Integration: The 22 × 80mm module uses an M.2 2280 Key-M connector with PCIe 3.0 x4 connectivity.
  • 2.5W Typical Power Consumption: Low typical power requirements support energy-efficient edge AI deployments.
  • On-Device Processing: Runs supported AI inference locally, reducing dependence on cloud services.
  • Cross-Platform Compatibility: Supports x86 and ARM architectures with Windows and Linux. Android support is currently available for B2B customers.
  • Framework and Model Support: Compatible with Keras, TensorFlow, TensorFlow Lite, PyTorch and ONNX, with access to over 150 pre-trained models through the online Model Zoo.
Brand ASUS
Product Name ASUS UGen300 M2 AI Module 2GB LPDDR4X
Manufacturer Model UGen300-M2-2G
SKU UGEN300-2GLD4X-M2
Product Type M.2 edge AI accelerator module
AI Processor Hailo-10H
AI Performance 40 TOPS at INT4 precision
20 TOPS at INT8 precision
Memory Capacity 2GB
Memory Type LPDDR4X
Memory Speed 4266 MT/s
Interface PCIe 3.0 x4
Form Factor M.2 2280
Connector Key Key-M
Typical Power Consumption 2.5W
Supported Host Architectures x86
ARM
Supported Operating Systems Windows
Linux
Android — currently available for B2B customers
Supported AI Frameworks Keras
TensorFlow
TensorFlow Lite
PyTorch
ONNX
Supported AI Model Categories Vision models
Generative AI models
Model Resources Over 150 pre-trained models through the online Model Zoo
Hardware Compatibility M.2 2280 Key-M interface with PCIe 3.0 x4 connectivity
Supported x86 or ARM host platform
Dimensions 22 × 80mm
Weight 6g

1 Year Walk-In Warranty

Ban Leong Technologies Limited
No. 150, Ubi Avenue 4, #04-01
Singapore 408825

Phone: +65 6512-9250
Email: support@banleong.com.sg
Opening times: 9:00am to 5:00pm (Monday - Friday) and 9:00am to 12:30pm (Saturday)

Warranty Claims are subject to Terms and Conditions*