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。

OneDiffusion

39 点作者 aarnphm超过 1 年前

2 条评论

fxtentacle超过 1 年前
In my opinion, this is way too &quot;abstract&quot;. Kinda like someone let ChatGPT run wild repackaging existing libraries with new APIs.<p>So this library abstracts away BentoML with a custom API. BentoML itself abstracts away its own internal framework. That inner framework abstracts away HuggingFace diffusers with a different API. And HF diffusers itself is a rather huge library that abstracts away using PyTorch to load the model and run it. HF diffusers contains lots of useful things for designing and training models, but none of that makes it through the 3 layers of API changes introduced here.<p>I predict that debugging any kind of production app using this is going to be pure hell. You&#x27;ll have 30 function deep call stacks across millions of lines of code. And that&#x27;s especially sad since the alternative is probably a single 300 LOC python script calling PyTorch directly.
评论 #37233164 未加载
twayt超过 1 年前
Thinly veiled attempt to advertise bentoml, their startup.<p>You can just use huggingface and Automatic111 for this, not sure why this is needed at all.
评论 #37233252 未加载
评论 #37233124 未加载
评论 #37233658 未加载