Login
Start Free Trial Are you a business? Click Here

Coral USB Accelerator Reviews

4.8 Rating 36 Reviews
Read The Pi Hut Reviews
Visit Product Page

Email:

contact@thepihut.com

David T
Verified Reviewer
Incentivized
Ordered this to speed up the AI functionality on my QNAP TS453BE NAS. It was instantly recognised by my NAS, although only when I used my own usb cable(!). The one supplied powers up the device, but it is not recognised. Other users have reported the same issue with the supplied USB cable. It has made a difference processing a library of nearly 40,000 photos; not a lighting fast improvement, but for the price I'm happy.
Helpful Report
Posted 1 month ago
Dan T
Verified Reviewer
Incentivized
Fantastic pr9duct. Using on QNAP NAS for AI services. Great proce and v good service asalways from Pi-Hut!
Helpful Report
Posted 2 months ago
Ricky B
Verified Reviewer
Incentivized
Super hard to get hold of these, and PiHut delivered quickly. great service.
Helpful Report
Posted 2 months ago
Sopuluchukwu S
Verified Reviewer
Incentivized
Came as described and arrived quite on time quality
Helpful Report
Posted 2 months ago
John E
Verified Reviewer
Incentivized
I intend to use the USB Coral TPU with the Frigate security camera NVR software under Linux but am not ready for doing that. So I researched how to verify the operation of the Coral using Linux. There are many articles online as to how to do this - most are out of date, or just plain don't work. However I was able to create a Docker container based on Debian 9 and Python 3.5 that worked with a Google demo. The Google software for the Coral requires a Python version below v3.9. What I did was use the method given here: https://www.jeffgeerling.com/blog/2023/testing-coral-tpu-accelerator-m2-or-pcie-docker but with my own Docker container created using the Dockerfile given below . As mine is the USB Coral I replaced "--device /dev/apex_0:/dev/apex_0" with "--device /dev/usb:/dev/usb" in the method. Dockerfile used: FROM debian:9 WORKDIR /home/john ENV HOME=/home/john RUN cd ~ RUN echo "deb http://archive.debian.org/debian stretch main" > /etc/apt/sources.list RUN echo "deb http://archive.debian.org/debian-security stretch/updates main" >> /etc/apt/sources.list RUN apt-get update --allow-insecure-repositories RUN apt-get install -y gnupg2 nano python3 pkg-config usbutils RUN apt-get install -y python3-dev RUN apt-get install -y python3-pil python3-numpy RUN apt-get install -y apt-transport-https RUN echo "deb https://packages.cloud.google.com/apt coral-edgetpu-stable main" \ | tee /etc/apt/sources.list.d/coral-edgetpu.list RUN apt-get install -y curl git wget RUN curl https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add - RUN apt-get update --allow-insecure-repositories RUN apt-get install -y edgetpu-examples RUN apt-get install -y gasket-dkms The output from the test software was: --------------------------- Poecile atricapillus (Black-capped Chickadee) Score : 0.441406 --------------------------- Poecile carolinensis (Carolina Chickadee) Score : 0.292969 So I assume this shows the Coral is working. The Coral was shipped from PiHut on the day of my order or the next, which was great.
Helpful Report
Posted 3 months ago
WARNING: Extremely difficult to set up. Even on standard Linux distributions like Ubuntu, you really need to check kernel module drivers support, their versions, and if you need to compile the driver from source, etc. It is far away from being plug and play. There's only one way to make it work in macOS, and it is by using QEMU.
Helpful Report
Posted 10 months ago
setting up was interesting but once done love it
1 Helpful Report
Posted 1 year ago
I bought this for Frigate in Home Assistant on my RPi5 with 5 cameras, and I have to say it now copes very well with object detection, reducing my main CPU usage massively as well as the system memory. I would definitely recommend this.
3 Helpful Report
Posted 1 year ago