NVIDIA Jetson Nano DevKit
472 GFLOPS edge AI — CUDA, TensorRT ready
IoT Kits
In StockGoogle Coral USB Accelerator
AI-CORAL-USB · Stock: 48
☆☆☆☆☆
No reviews yet — be the first
₹4,979
₹6,639
incl. GST
[Raspberry Pi] [Arduino]
Google Coral USB Accelerator adds an Edge TPU coprocessor delivering 4 trillion operations per second for TensorFlow Lite models. Plug into any Linux host including Raspberry Pi.
3 clicks: Buy Now → Checkout → Place Order
Edge TPU — 4 TOPS ML inference over USB
High-res visual shown left · QA bench-tested before dispatch.
| Operating Voltage | USB 5V |
|---|---|
| Peak Perf | 4 TOPS |
| Interface | USB 3.0 Type-C |
| Supported | TensorFlow Lite |
| Power | 1W idle / 2.5W peak |
| Dimensions | 65 × 30 × 8 mm |
| Weight | 20g |
USB-C
Host connection
Status LED
Power/Activity
Wiring guide: power rails first (VCC/GND), then bus (I2C/SPI), then interrupt/PWM lines.
// Coral runs on host Linux — MCU sends sensor data
void setup() { Serial.begin(115200); }
void loop() {
Serial.println(analogRead(A0));
delay(100);
}
from pycoral.adapters import common, detect
from pycoral.utils.edgetpu import make_interpreter
from PIL import Image
interpreter = make_interpreter('model_edgetpu.tflite')
interpreter.allocate_tensors()
image = Image.open('frame.jpg')
common.set_input(interpreter, image)
interpreter.invoke()
objs = detect.get_objects(interpreter, score_threshold=0.4)
print(objs)
// Coral runs on host Linux — MCU sends sensor data
void setup() { Serial.begin(115200); }
void loop() {
Serial.println(analogRead(A0));
delay(100);
}
Technical datasheet for AI-CORAL-USB.
Community signal
Ratings & Reviews
Real maker feedback for Google Coral USB Accelerator.
—
No reviews yet — be the first
No public reviews yet
Share a wiring tip, build note, or bench-test result to help the next maker.