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。

A brief history of LLaMA models

245 点作者 andrewon大约 2 年前

8 条评论

jiggawatts大约 2 年前
It keeps saying the phrase “model you can run locally”, but despite days of trying, I failed to compile any of the GitHub repos associated with these models.<p>None of the Python dependencies are strongly versioned, and “something” happened to the CUDA compatibility of one of them about a month ago. The original developers “got lucky” but now nobody else can compile this stuff.<p>After years of using only C# and Rust, both of which have sane package managers with semantic versioning, lock files, reproducible builds, and even SHA checksums the Python package ecosystem looks ridiculously immature and even childish.<p>Seriously, can anyone here build a docker image for running these models on CUDA? I think right now it’s borderline impossible, but I’d be happy to be corrected…
评论 #35758763 未加载
评论 #35761453 未加载
评论 #35758057 未加载
评论 #35759320 未加载
评论 #35758636 未加载
评论 #35758006 未加载
评论 #35759766 未加载
评论 #35757903 未加载
评论 #35757987 未加载
评论 #35757917 未加载
评论 #35759923 未加载
doodlesdev大约 2 年前
<p><pre><code> &gt; Our system thinks you might be a robot! We&#x27;re really sorry about this, but it&#x27;s getting harder and harder to tell the difference between humans and bots these days. </code></pre> Yeah, fuck you too. Come on, really, why put this in front of a _blog post_? Is it that hard to keep up with the bot requests when serving a static page?
评论 #35758552 未加载
vessenes大约 2 年前
Most places that recommend llama.cpp for mac fail to mention <a href="https:&#x2F;&#x2F;github.com&#x2F;jankais3r&#x2F;LLaMA_MPS">https:&#x2F;&#x2F;github.com&#x2F;jankais3r&#x2F;LLaMA_MPS</a>, which runs unquantized 7b and 13b models on the M1&#x2F;M2 GPU directly. It&#x27;s slightly slower, (not a lot), and significantly lower energy usage. To me the win not having to quantize while not melting a hole in my lap is huge; I wish more people knew about it.
评论 #35759431 未加载
评论 #35761498 未加载
simonw大约 2 年前
I&#x27;m running Vicuna (a LLaMA variant) on my iPhone right now. <a href="https:&#x2F;&#x2F;twitter.com&#x2F;simonw&#x2F;status&#x2F;1652358994214928384" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;simonw&#x2F;status&#x2F;1652358994214928384</a><p>The same team that built that iPhone app - MLC - also got Vicuna running directly in a web browser using Web GPU: <a href="https:&#x2F;&#x2F;simonwillison.net&#x2F;2023&#x2F;Apr&#x2F;16&#x2F;web-llm&#x2F;" rel="nofollow">https:&#x2F;&#x2F;simonwillison.net&#x2F;2023&#x2F;Apr&#x2F;16&#x2F;web-llm&#x2F;</a>
评论 #35757088 未加载
brucethemoose2大约 2 年前
There is also CodyCapybara (7B finetuned on code competitions), the &quot;uncensored&quot; Vicuna, OpenAssistant 13B (which is said to be very good), various non English tunes, medalpaca... the release pace maddening.
评论 #35757567 未加载
brianjking大约 2 年前
I&#x27;ll never understand why everyone is spending so much time on a model you cannot use commercially (at all).<p>Secondly, most of us can&#x27;t even use the model for research or personal use, given the license.
评论 #35759832 未加载
评论 #35761038 未加载
评论 #35760510 未加载
评论 #35759805 未加载
评论 #35761968 未加载
评论 #35759714 未加载
评论 #35761081 未加载
FloatArtifact大约 2 年前
There needs to be a slight dedicated to tracking all these models with regular updates.
评论 #35758393 未加载
foobarbecue大约 2 年前
Ok I gotta know... what&#x27;s the art?