Coral Dev Board Mini
The cost-effective development board option for the EdgeTPU platform
A complete development platform for fast inferencing
The Coral Dev Board Mini is an ARM-based single-board computer (SBC) development board that includes a specialized Coral Edge TPU for acceleration of machine learning (ML) inferencing calculations.
The Edge TPU coprocessor is capable of 4 trillion operations per second, using only 2 Watts of power.
Supports Mendel Linux (a derivative of Debian Linux).
Simply compile TensorFlow Lite models to run on the Edge TPU. Also supports AutoML Vision Edge: easily build and deploy fast, high-accuracy custom image classification models to your device.
- MediaTek 8167 SoC
- Quad-Core ARM Cortex-A35, up to 1.5 GHz, 64-bit ARMv8-A
- Imagination PowerVR GE8300 GPU (supports Vulkan 1.0)
- 1080p/60fps HEVC & MPEG4 decoder
- Google Edge TPU coprocessor
- 4 TOPS (int8); 2 TOPS per watt
- connected to MediaTek 8167 SoC via USB 2.0 interface
The on-device ML processing reduces latency and avoids dependence on a stable network connection. It is ideal for image classification processing tasks.
Note: for optimal performance, we recommend the Coral Dev Board 4 GB, or the Coral USB Accelerator, which have higher-speed interfaces to the Edge TPU.
- 2 GB LPDDR3
- 8 GB eMMC
Interfaces & Connectivity
- Wi-Fi 5 (802.11a/b/g/n/ac, 2.4/5GHz) (MediaTek M7658)
- Bluetooth 5.0 & Bluetooth Low Energy (BLE)
- 1 x USB type-C (USB 2.0) OTG for data, can operate as host and as peripheral
- 1 x USB-type C for power
- 3.5 mm audio jack
- digital PDM microphone
- 2.54mm 2-pin mono speaker terminal
- microHDMI (v1.4), up to 1920x1080 resolutions
- 24-pin FFC connector for MIPI-CSI2 camera (4-lane), up to 8 MP camera supported
- 24-pin FFC connector for MIPI-DSI display (4-lane)
- microSD slot, supports SD/SDIO 3.0, max 32 GiB card capacity
- 40-pin GPIO header (3.3 V voltage levels)
Refer to the datasheet for a GPIO pinout. The board supports:
- Mendel Linux (Debian Linux derivate)
- supports TensorFlow Lite models to run on it's Edge TPU
- supports AutoML Vision Edge, for fast deployment of image classification models
- Supports root of trust with on-board cryptochip A71CH
- ARM TrustZone for chip-to-cloud security
- power button
- programmable user button
- requires 5 V / 2 A USB-C power supply
The Coral Dev Board Mini comes pre-flashed with Mendel Linux on the eMMC memory. It includes all the software you need to run TensorFlow Lite models on the Edge TPU.
Dimensions: 64 mm x 48 mm x 14.6 mm
Weight: 25.5 g
Google part number: G650-03324-01
Asus part number: 90AN00X0-B0XAY0
Google model number: KA1
Datasheet & Benchmarks
Additional Legal Information
The products offered by Google are unrelated to the products offered under the CORAL trademarks owned by Orient Development Enterprises Ltd. in Taiwan.