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: Replacing Nvidia GPU with AMD

19 点作者 textcortex超过 1 年前
Dear community I am currently exploring the feasibility of replacing Nvidia GPUs with AMD GPUs for LLM inference. Can anyone share their experience or point me to any relevant research on the performance differences between these GPUs for this task? Are there any particular hardware or software limitations that may affect the feasibility of such a switch? Thank you for your insights!

6 条评论

cjbprime超过 1 年前
Since the amount of VRAM is low (max 24GB), one question to look into before investing might be whether there's support for chaining multiple cards together.
评论 #38762129 未加载
LorenDB超过 1 年前
Ollama just had AMD support merged. I haven't got it working with my 6700 XT eGPU yet, but I anticipate getting there soon.
评论 #38762099 未加载
htrp超过 1 年前
You'll be fine for inference but probably struggle to run any large models requiring multi gpu ops.
InitEnabler超过 1 年前
I&#x27;d take a look ROCm docs: <a href="https:&#x2F;&#x2F;www.amd.com&#x2F;en&#x2F;developer&#x2F;resources&#x2F;rocm-hub.html" rel="nofollow noreferrer">https:&#x2F;&#x2F;www.amd.com&#x2F;en&#x2F;developer&#x2F;resources&#x2F;rocm-hub.html</a>
评论 #38758157 未加载
Havoc超过 1 年前
My understanding is inference on latest amds (79*) is mostly fine but training is still shaky
brudgers超过 1 年前
Is there a clear business rationale?<p>Can that rationale be mitigated by increasing revenue as an alternative to adding the technical risks implicit in your question?<p>I mean it is one thing if this is a hobby project and using AMD will provide interesting challenges to keep yourself occupied. How you spend your own time and money is entirely up to you.<p>But spending other people’s money is another thing entirely (unless they are mom and dad). And even more so spending other people’s time particularly when it comes to paychecks.<p>Finally, I am not saying there aren’t business cases where AMD doesn’t make sense. Just that Nvidia is the default for good reasons and is the simplest thing that might work. Good luck.