Passes the "woman on grass" test ; - )<p>Seriously though, there are some minor hand issues and a rare missing body part. "Correct anatomy, no missing body parts." seems to fix it mostly. Still pretty good for an early 0.1 announcement.<p>Following full sentences is pretty good. Although this: "A photo of a table. On the table there's a green box on the right, a red ball on the left. There's a yellow cone on the box." keeps putting the cone on the table.<p>Not trained on naked bodies though - generates blob monsters instead.
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://ideogram.ai/g/ENMWd7PrQ32dIWSF91uMJQ/2" rel="nofollow">https://ideogram.ai/g/ENMWd7PrQ32dIWSF91uMJQ/2</a> comes out exposing that training didn't have enough naked bodies. Prompt adherence is very very good otherwise. Can try top images of the day/hour from ideogram to test.
In case you missed it, the authors were pretty smart to include that folded section in the middle, "Prompt for prompt-enhancement". I slapped that into gpt (<a href="https://chatgpt.com/share/2e53403e-4bd7-4138-ac34-55378e2ed306" rel="nofollow">https://chatgpt.com/share/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'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.
AIs are still not able to understand negations.<p>Try "ramen without egg" or "ramen with no egg" and it will show ramen WITH egg.<p>Or "man without striped shirt" will give "man WITH striped shirt"
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?
So, now that this is released are we no longer going to have pendant people complaining that this "isn't real open source"?<p>Here is your model, complainers.<p>I'm not really sure why you'd be so insistent on that, as opposed to just fine tuning the "totally not open source, but instead just open weights" 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.