G950-01456-01/ G950-06809-01Coral USB Edge TPU ML Accelerator coprocessor for Raspberry Pi and Other Embedded Single Board Computers

USB Accelerator

Important Note: This product includes the USB Accelerator only (Model: G950-01456-01 / G950-06809-01).

AI for Everyone

Bring real-time ML inference to your existing hardware! The USB Accelerator features a powerful Edge TPU coprocessor connected via USB 3.0, delivering fast, energy-efficient inference for TensorFlow Lite models.

Key Benefits:

• Local Processing Data stays on-device for low latency and full GDPR compliance
• High Performance Up to 4 trillion operations per second at just 2W power consumption
• Easy Integration Works with Raspberry Pi 4, Linux, macOS, and Windows systems

Perfect for Raspberry Pi 4

Achieve up to 20x faster inference compared to Pi 4 alone. Real-time video recognition at 50+ fps becomes possibleunattainable without the accelerator.

Technical Specifications

| Feature | Specification |

| -------------- | ------------------------------- |

| Processor | Edge TPU ML accelerator |

| Interface | USB 3.0 (Type-C) |

| Compatibility | Linux, macOS 10.15+, Windows 10 |

| Power | 900 mA peak @ 5V |

| Dimensions | 65 × 30 × 8 mm |

| Python Support | 3.5, 3.6, 3.7 |

System Requirements

• Linux Debian 6.0+ or derivatives (Ubuntu 10.0+, Raspbian)
• Architecture: x86-64, ARMv7 (32-bit), or ARMv8 (64-bit)
• macOS 10.15 with MacPorts or Homebrew
• Windows 10
• USB 3.0 port recommended
• Python 3.53.7

Package Contents

• USB Accelerator
• USB 3 cable

Operating Temperature

• 25°C Maximum clock frequency (optimal performance)
• 35°C Reduced clock frequency

Caution: Device may become hot during operation. Allow cooling before handling. Operation outside recommended temperature range is at user's own risk.

Part Number: G950-01456-01 / G950-06809-01 (from Oct 2020)

ASUS Part Number: 90AN0020-B0XAY0