![](https://i.imgur.com/3lYP3BY.png)

<center>

# Stable Whaaaaat?!

</center>

Yes I promise, it is stable.. ish

![](https://i.imgur.com/BfbUfBe.png)

<hr>

<h2>Artificial Intelegence?</h2>

<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AL1cA47guD9PC5LL5YQtBtWhPzHRqbxu8eUEYcW3LGySoMVkiHoejkDiRjtUVvX.jpg" alt="a fantasy landscape featuring mystical plants  area  landscape  hyper realistic  detailed  fantasy  dramatic  cinematic [Stable Diffusion plms 2.png">

So after some playing around and testing over a couple of weeks I think I may have (somewhat) found a way to more or less get what I want out of the AI.

<div class="pull-left"><div class="text-justify">
<h3>The Fun Stuff</h3>
So as you start to understand(more or less) how the diffusion and text models interpret your words it will all come together and you can start writing prompts and tweak some settings to make some super cool things...
<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKYLP9wj3XrsD3L9bKpKAtBSb1nTKtMcme8hxHgzAJHdSyYPZsGsgWneMFa9pKf.jpg" alt="a fantasy landscape in the woods featuring mystical plants  area  landscape  hyper realistic  detailed  fantasy  dramatic  cinematic [Stable D 3.png">


</div></div>
<div class="pull-right"><div class="text-justify">
<h3>Not So Fun Stuff</h3>
While playing around with a lot of settings and models it can get quite annoying and tiring when the produced result is just a tinnyyy bit off. Which leads to some wacky creations...



<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKLt4ieR6R8g78PxNS5PMarnUY7giETCTLHP1qDYWYsfnZJCuvA2ruEP9PvHsSY.png" alt="cute chibi animal  digital art  mascot  chibi  cute  adorable [Stable Diffusion plms] 361281737.png">


</div></div>
<hr>
<h1>Whatcha usiiing?</h1>

<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKDhF6YJwdrEoSGg5WZtuRj2ixQML32tchVAkaxbqPRFTFvbCuw58oMeNDtdgNN.png" alt="a galaxy in a small glass bottle as a 3d render  dreamy  mystical  chaotic  disturbed [Stable Diffusion plms] 1352883676.png">

As many of my previous AI posts, I use  Visions Of Chaos with a bunch of AI models; for text to image generation I typically use StableDiffusion but in the past I had used VQGan of DALL-E.

<hr>

<h2>Okay, but gimme the settings bro...</h2>

Well, that is easier said than done, for the most part its not just a one-size fits all solution, its more a duct-tape kind of solution (for now). But here are some tips:
<hr>
<div class="pull-left"><div class="text-justify">
<h3>Big Yesses</h3>
For the most part, any prompt will create something more or less decent, but just like regular art you need to carve it out. When giving it prompt try to be as specific as you want, you want the image to look like a digital painting, cinematic, have a dark ambience etc. - then specify it.

</div></div>
<div class="pull-right"><div class="text-justify">
<h3>Big NO NO's</h3>
AI is not some magical tool that just knows what you want it produce, it can only do what you tell it to do. So the more vague you are the more it will struggle and the results will look poor.
</div></div>

<hr>
<center>
<h4>Some Examples</h4>
The images below are prime examples, the images on the left showcase barely any descriptors while the images on the right contain far more descriptiors. All the artworks are generated over 500 iterations <br><br>*NOTE FOR THE FIRST TWO I FORGOT TO NOTE THE SEED*
</center>

<hr>

<div class="pull-left"><div class="text-justify">
<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKF7Btf7w5qgYHVuayCr1XryBwGMRAwMCNfsuLi5hMikXB2X8Af4A3nry6CyzKz.jpg" alt="a fantasy landscape [Stable Diffusion plms] 1219434913.png">
<sub><pre>a fantasy landscape</pre></sub>
<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKAoqHrG4imyjN4aPrC3NvJsLfcGuF9jFdYPEs11FmUFNHo3Qi1HRQiHiJDryES.png" alt=" thresh combined with pyke from league of legends hooking a soul with his hook [Stable Diffusion plms] 296945706.png">
<sub><pre>thresh combined with pyke from league of legends hooking a soul with his hook
seed:296945706</pre></sub> 
</div></div>
<div class="pull-right"><div class="text-justify">
<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKCBqgAZvx4ZPBLLL7KQ3HmCPdBks1RgGhnV54h5ig91rL1EDrvAT8TvEivvsKd.jpg" alt="a fantasy landscape featuring mystical plants  area  landscape  hyper realistic  detailed  fantasy  dramatic  cinematic [Stable Diffusion plms 3.png">
<sub><pre>a fantasy landscape featuring mystical plants  area  landscape  hyper realistic  detailed  fantasy  dramatic  cinematic</pre></sub>
<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AKGYUgQUsPeEAqBDEoPuQhvDcPVUtb6k9EUB2MagAFc2Nm81SpVsB3wAEgbgEAm.png" alt=" thresh combined with pyke hooking a soul from league of legends  league of legends  detailed  cinematic  dark  digital artwork  concept art   11.png">
<sub>
<pre>thresh combined with pyke hooking a soul from league of legends, league of legends, detailed, cinematic, dark, digital artwork, concept art, ghostly hands holding the hook, armoured skeleton, thresh, pyke,cinematic lighting, highly detailed, illustration, artstation, digital painting, art by  Bob Kehl, Terry Moore, and Greg Rutkowski
seed: 1664105712
</pre>
</sub> 
</div></div>

<hr>
<h2>Okay then..</h2>
Yep, as you can see adding more details and descriptors can really take an image to the next level. But ofcourse as it stands, AI is a tool not a magic one fit solution.

As for my settings, in VoC this would be it: 

<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/23sxmuZ9cPRYExjeSzDbocNGJYZCp2qyviDJBbte8LtnhEX4xQNiAXbczgTk9ctrBMENn.png" alt="image.png">

I just have quite a lot of HD passes on the image to take it from a blurry 512x512 to a nice crisp and detail 4096x4096

<hr>
<h1>Conclusion</h1>

After using VQGAN and a couple of other models such as disco I can definitely say StableDiffusion is the current best model for getting more natural and coherent results.. for the most part sometimes you are left questioning where you went wrong like this:

<img src="https://files.peakd.com/file/peakd-hive/deadeagle63/AK6YNUGTCjWpj8F8pcy4KctpitrkAqstDptkTerLDps2ckwSy5Egv5imBFGNRoX.png" alt="image.png">
<sub>What?!</sub>
<hr>
<h1>Deadeagle63</h1>