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。

AuraFlow v0.1: a open source alternative to Stable Diffusion 3

164 点作者 treesciencebot10 个月前

7 条评论

viraptor10 个月前
Passes the &quot;woman on grass&quot; test ; - )<p>Seriously though, there are some minor hand issues and a rare missing body part. &quot;Correct anatomy, no missing body parts.&quot; seems to fix it mostly. Still pretty good for an early 0.1 announcement.<p>Following full sentences is pretty good. Although this: &quot;A photo of a table. On the table there&#x27;s a green box on the right, a red ball on the left. There&#x27;s a yellow cone on the box.&quot; keeps putting the cone on the table.<p>Not trained on naked bodies though - generates blob monsters instead.
评论 #40942785 未加载
评论 #40942939 未加载
smusamashah10 个月前
Prompt adherence is great. I copied a few prompts from ideogram (which also adheres to prompt) and results were good until they involve female bodies. This for example <a href="https:&#x2F;&#x2F;ideogram.ai&#x2F;g&#x2F;ENMWd7PrQ32dIWSF91uMJQ&#x2F;2" rel="nofollow">https:&#x2F;&#x2F;ideogram.ai&#x2F;g&#x2F;ENMWd7PrQ32dIWSF91uMJQ&#x2F;2</a> comes out exposing that training didn&#x27;t have enough naked bodies. Prompt adherence is very very good otherwise. Can try top images of the day&#x2F;hour from ideogram to test.
评论 #40943917 未加载
halr900010 个月前
In case you missed it, the authors were pretty smart to include that folded section in the middle, &quot;Prompt for prompt-enhancement&quot;. I slapped that into gpt (<a href="https:&#x2F;&#x2F;chatgpt.com&#x2F;share&#x2F;2e53403e-4bd7-4138-ac34-55378e2ed306" rel="nofollow">https:&#x2F;&#x2F;chatgpt.com&#x2F;share&#x2F;2e53403e-4bd7-4138-ac34-55378e2ed3...</a>) and made a few prompts. Ran those on their online demo. Initial impressions:<p><pre><code> - prompt adherence is really good - it&#x27;s somewhere between SD15 and SDXL at creating pictures of text - aesthetic quality is good, but leaves some to be desired </code></pre> Gonna play more with it in ComfyUI.
executesorder6610 个月前
AIs are still not able to understand negations.<p>Try &quot;ramen without egg&quot; or &quot;ramen with no egg&quot; and it will show ramen WITH egg.<p>Or &quot;man without striped shirt&quot; will give &quot;man WITH striped shirt&quot;
评论 #40944018 未加载
评论 #40943970 未加载
评论 #40944694 未加载
skybrian10 个月前
Fails on “piano keyboard” (shows a full piano) and “close up of piano keyboard,” (bizarre duplicate keyboard monstrosity.)<p>It’s a difficult prompt. Nobody gets the grouping of black keys right. Maybe someday?
gorkemyurt10 个月前
<a href="https:&#x2F;&#x2F;x.com&#x2F;ostrisai&#x2F;status&#x2F;1811620901420429441" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;ostrisai&#x2F;status&#x2F;1811620901420429441</a>
评论 #40943389 未加载
stale200210 个月前
So, now that this is released are we no longer going to have pendant people complaining that this &quot;isn&#x27;t real open source&quot;?<p>Here is your model, complainers.<p>I&#x27;m not really sure why you&#x27;d be so insistent on that, as opposed to just fine tuning the &quot;totally not open source, but instead just open weights&quot; models.<p>But go ahead, I guess.<p>Now we can get back to talking about capabilities, usage, and results, as opposed to arguing about the definition of words.