TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Coral USB Accelerator with Google's Edge TPU

75 点作者 achristmascarl3 个月前

16 条评论

geerlingguy3 个月前
These were nice early in the TensorFlow evolution, for things like Frigate...<p>But even CPU inference is both faster and more energy efficient with a modern Arm SBC chip, and things like the Hailo chip are way faster for similar price, if you have an M.2 slot.<p>I haven&#x27;t seen a good USB port alternative for edge devices though.<p>The big problem is Google seems to have let the whole thing stagnate since like 2019. They could have some near little 5&#x2F;10&#x2F;20 TOPS NPUs for cheap if they had continued developing this hardware ecosystem :(
评论 #43122872 未加载
评论 #43133020 未加载
评论 #43122887 未加载
评论 #43123337 未加载
评论 #43123754 未加载
jasongill3 个月前
Not sure if there is something new here but it looks like the same product that has been around for a few years now (wasn&#x27;t Coral released in 2019-ish?)
评论 #43121765 未加载
评论 #43122502 未加载
franciscop3 个月前
I have a couple of these, unfortunately I&#x27;ve been waiting for the ecosystem to get better and run newer&#x2F;improve models to no avail. I attempted some YOLO ports (since Coral uses a specific architecture) and not sure if I&#x27;m just bad at this or it&#x27;s actually hard, but beyond the basic examples with Google&#x27;s own ecosystem I wasn&#x27;t able to run anything else on these. I was hoping an upgrade from seeing this on HN, but it seems to be the same old one.
评论 #43123871 未加载
评论 #43122866 未加载
评论 #43123271 未加载
评论 #43122621 未加载
milesward3 个月前
I put 8 of these on a mini-ITX computer on a fake moon in my backyard, AMA.
评论 #43122160 未加载
评论 #43122465 未加载
评论 #43122714 未加载
评论 #43122691 未加载
评论 #43122485 未加载
评论 #43123109 未加载
评论 #43122566 未加载
ein0p3 个月前
This is way too outdated to be relevant in any way. Back in the day they had a board with a TPU on it, before everyone else did. That board ran object detection at a pretty good resolution at like 80fps in 2.5W power budget. I still have that board in my drawer - I never did find any use for it at its price point. Plus, because it&#x27;s Google, I expected that they&#x27;d abandon the board within 2 years tops, which is exactly what happened. The board was like $100 IIRC which was a good chunk of cash when RaspberryPi was like $25. Nowadays there are _dozens_ of Chinese boards available with on-chip TPUs. Tooling still sucks mightily, but that&#x27;s expected when dealing with embedded systems. Unlike with the Google board, you can usually build your own Linux for these using Yocto or Buildroot with minimal tweaks.
评论 #43128815 未加载
thedougd3 个月前
FWIW, The M.2 and mini PCIe form factors are more cost effective. I added one to the WiFi+Bluetooth slot on a refurbished Dell desktop to perform object detection in my CCTV NVR.<p><a href="https:&#x2F;&#x2F;frigate.video&#x2F;" rel="nofollow">https:&#x2F;&#x2F;frigate.video&#x2F;</a>
评论 #43122008 未加载
runjake3 个月前
I have one of these. The USB model sucks. It overheats unless you put them in high efficiency (low performance) mode which defeats the purpose.<p>The mini-PCIe variant is much more reliable, but I ended up ditching the Coral entirely and replacing it with a GTX 1060.
userbinator3 个月前
USB <i>AI</i> accelerator, in case anyone was wondering why USB needed to be accelerated.
muxamilian3 个月前
It seems more than dead and only supports small neural networks. Viable alternatives are Hailo and Axelera (<a href="https:&#x2F;&#x2F;www.axelera.ai" rel="nofollow">https:&#x2F;&#x2F;www.axelera.ai</a>), which is a newer.
qwe----33 个月前
Maybe I’m Stupid but I couldn’t figure out how to set the pull up or pull down resistors on these boards. Maybe with LLMs I can figure this out now…. Something to do with device tree confits?
bdhcuidbebe3 个月前
What about the Coral? I been running frigate with mine for 2 years.
评论 #43121954 未加载
metadat3 个月前
How does a Coral unit compare to a current gen AMD or Intel CPU in terms of throughput for ML tasks?
评论 #43122679 未加载
tonymet3 个月前
a couple questions for people who have been using it. Where does this fit between a typical budget arm cortex and a gpu? and what are the practical sized models you could run on one of these?
fitsumbelay3 个月前
... and one day I&#x27;ll get mine to work ...
mewmix3 个月前
copyright 2020 on their site lol.
mewmix3 个月前
copyright 2020 on their site lol