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。

Ask HN: Rate GPUs

1 点作者 homedepotdave超过 3 年前
As a deep learning hobbyist, I have a general sense for which GPUs are great and which are awful. But, I&#x27;m less familiar with some of the in-the-middle ones, and I&#x27;m generally curious what criteria pro researchers use. E.g., never use anything besides X, if you were strapped for cash maybe use Y, etc.<p>Here are the main ones I&#x27;ve come across, but maybe people would say others are relevant too.<p>I&#x27;ve included my guesses as to ratings&#x2F;thoughts, as well as current AWS spot price for 1 GPU in east-2. Does this assessment seem roughly right? Am I missing anything? Obviously assessing these can be very complicated based on what exactly you&#x27;re using them for; I&#x27;m just looking for very high level thoughts.<p>A100 9.5&#x2F;10 - among the best available now? $1.39&#x2F;hr (this is an estimate; AWS actually only offers an 8 GPU instance for $11.15&#x2F;hr so I just divided by 8)<p>V100 8.5&#x2F;10 - very good $0.92&#x2F;hr<p>P100 8&#x2F;10 - good, used a fair amount in research AWS doesn&#x27;t seem to offer these anymore? Replaced by V100s?<p>T4 5&#x2F;10 - I don&#x27;t know. Do people use these ever? $0.36&#x2F;hr<p>M60 2&#x2F;10 - I don&#x27;t really know. I assume bad $0.34&#x2F;hr<p>K80 1&#x2F;10 - crap, basically don&#x27;t ever waste your time $0.27&#x2F;hr

1 comment

harias超过 3 年前
Slightly outdated guide: <a href="https:&#x2F;&#x2F;timdettmers.com&#x2F;2020&#x2F;09&#x2F;07&#x2F;which-gpu-for-deep-learning&#x2F;" rel="nofollow">https:&#x2F;&#x2F;timdettmers.com&#x2F;2020&#x2F;09&#x2F;07&#x2F;which-gpu-for-deep-learni...</a><p>See Also: <a href="https:&#x2F;&#x2F;lambdalabs.com&#x2F;gpu-benchmarks" rel="nofollow">https:&#x2F;&#x2F;lambdalabs.com&#x2F;gpu-benchmarks</a>