Playground Introduction To Canvas

Playground Introduction To Canvas. Hello everyone! This is Part 1 of a series of videos that will be producing to get you all familiar with Canvas. This video is an overview of some of the main features but you can expect that we will be diving deeper in videos to come!
More in-depth videos on related topics
Prompting 101 • Playground AI tutorial...
Samplers • What's The Difference ...
Seeds • Playground AI Tutorial...
ControlNet • Getting Started With C...
0:00 Generating your first image
0:56 Navigating canvas
1:20 Introduction to filters
2:38 Image dimensions
3:40 Prompt Guidance, Quality & Details
6:26 Introduction to Seeds
9:07 Introduction to samplers
11:19 Secondary Menu, Crop tool
11:52 Introduction to Image to image
12:51 Basic compositing
14:29 Upscaling
14:56 Downloading your images
15:14 Information panel
15:56 Intro to Outpainting
16:53 Object Eraser
18:38 Publishing your image
Find us on
Playground playgroundai.com
Discord / discord
Twitter / playground_ai

[@playgroundai]

Hope you enjoyed Part 1! Let us know in the comments what you want us to cover in future videos!

[@MakeTechPtyLtd]

looks great. Just using Canvas with infill right now. I thought I saw it was SD XL in Canvas mode but actually I can't see the model on this page [edit; I see SD XL when clicking on an image] . In any case, it works great. Sometimes when trying to generate an infill it does a new layer image instead. Dreambooth filters in in-painting would be cool, but maybe SDXL doesn't need it because of its 2 mode architecture? Id like to ise the dream both models in in-painting

[@MissChelle]

I would like to see a video just for us iPad users that don’t have a keyboard to use control shortcuts. Pretty please 😊 🇦🇺❤

[@kemikemi756]

Thanks !! Will stay tuned for moar

[@camo4012]

It's fantastic WHEN YOU IMPROVE THE GENERATIVE FILLING, ufffff BEING CRAZY VERY GOOD THE BEST TOOL, WHEN YOUR GENERATIVE FILLING IS LIKE THE ONE IN PHOTOSHOP BETA WITHOUT A DOUBT YOU WILL HAVE THE BEST WEB PAGE

[@PolycultureArt]

I would like to know the benefit of the generation frame covering the image you select for image to image. I dont want to copy over the old image I want to generate next to it. I'm sure there must be a reason for this UI behaviour, but I dont really get it..

[@vishalpisote1306]

This is completely Amazing. Playground seems to be the most sorted and simple AI. It saves time and efforts to generate images. Thank you for this amazing innovation.

[@playgroundai]

We're glad you are happy with it! More to come!

[@supersouls6480]

Awesome tutorial and great technology, can see how much hard work was put into it. Thank you for providing it for free

[@playgroundai]

You’re welcome! Glad your enjoying the product. 👍🏼

[@jordaanrob]

This is out of this world. Playground is truly amazing in every way.

[@playgroundai]

Glad to hear! Enjoy playground!

[@MauriceNaragon]

I don't know if I have ever written a comment on a KZbin tutorial or video in general, but I felt I needed to do so here. Your presentation style is wonderful and very informative. I really enjoyed learning more through this tutorial. Great tools, great presentation. Thank you!

[@playgroundai]

Hey Maurice, such an encouraging comment, thank you for taking the time to do so! I hope I can continue to deliver the same value through these videos. 👍🏼

[@umarsidiki2277]

Thank you for the detailed video. Thank you for keeping the tool almost free :) Thank you for normalizing technology.

[@playgroundai]

Our pleasure! And thank you for the support! 🙌

[@TheGameIsOverCy]

This video was helpful and informative. I learned a lot about Canvas and its features. I especially liked how you explained the different tools and how to use them. Thank you for making this video and I look forward to the next one.

[@innalugo1363]

I love this tool, it is complete and at the same time simple, easy to use, it is incredible. The creators are geniuses and although I have tried others I always return to it. That's why I love her. Thank you

[@playgroundai]

You're welcome and thank you for the support! 👍

[@andreads502]

I'm amazed by all the possibilities, great tutorial video btw. Thanks!!

[@playgroundai]

You're welcome! There really is so much you can do!

[@SUDHANSHU934]

I just started out of curiosity and then playground ai becomes my go to tool for everything around image...... awesome video....Thanks for the efforts you are putting out here.

[@playgroundai]

You’re welcome and thank you for the support 🙌🏼

[@RhillEnglish]

🎯 Puntos clave para una navegación rápida: 00:00 🎬 El video es una introducción a Canvas, parte de una serie que continuará en semanas posteriores. 00:26 🖼️ Canvas es un lienzo donde se generan imágenes a partir de descripciones conocidas como "prompts". 01:20 🎨 Puedes aplicar filtros predefinidos a tus imágenes en Canvas para darles diferentes estilos. 02:31 🗑️ Puedes eliminar imágenes en Canvas si no te gustan o si te equivocas, y también puedes deshacer acciones con Ctrl+Z. 03:26 📏 Puedes cambiar las dimensiones de las imágenes y ajustar el tamaño del cuadro de generación en Canvas. 05:16 📊 Ajustar la "prompt guidance" y la "quality and details" en Canvas afecta la calidad y el estilo de la imagen generada. 08:30 🌈 Puedes cambiar colores, ropa y detalles de una imagen en Canvas ajustando la descripción (prompt). 09:12 🤖 En Canvas, puedes utilizar diferentes "samplers" para reducir el ruido y ajustar la variación en las imágenes generadas. 11:13 ✂️ Canvas ofrece herramientas para recortar imágenes, ajustar la fuerza de las variaciones y eliminar objetos no deseados. 15:05 🖌️ La función "out paint" en Canvas te permite extender una imagen y realizar ediciones finales como ajustes de brillo, contraste y saturación. 18:49 📤 Puedes publicar tus imágenes en Canvas, agregar una descripción y decidir si deseas que sean públicas o privadas en tu galería de perfil.

[@heidialvarado6409]

Thank you so much for explaining everything. I really appreciate it. I love playground!! 👍🤗

[@alchemydelights6406]

Great tutorial! I too, loved your presentation style - concise and to the point. These tutorials make me appreciate Playground Ai even more. Keep it up!!

[@playgroundai]

Thanks so much! That means a lot to me! More to come!

[@PolycultureArt]

Cant wait for an advanced canvas video :) I'm sure there are tips and tricks im missing.. Love you guys!

[@playgroundai]

Coming soon!

[@tala77700]

this feels like magic !! I am so glad I found this , thank u very much !!

[@playgroundai]

You're very welcome!

[@johnhawkinshawkins1284]

at last! this is the most comprehensive AI tutorial around. Thank you.

[@playgroundai]

You’re welcome! We’ll have part 2 soon!

[@rasterman2792]

man i love you explain everything to detail great work

[@playgroundai]

We appreciate it! I'm glad that it was helpful for you. It's important to understand the basics to move on to the next videos so you are one step ahead! Thanks for the support!

[@Wheelly1]

Your app is amazing. I see some perfectionism coming from it. It`s a real pleasure to use it. Thank you so much!

[@playgroundai]

Very kind of you thank you! We're always working on making it better and improving it, more to come!

[@user-rm7sq6kk4v]

I love your presentation, clear, good pace, and easy to understandable instruction. I will be watching more to understand the use of Playground AI. Thanks

[@playgroundai]

Thank you for taking the time to leave some feedback. I take pride in my presentation and I constantly try to improve, so thank you! Please let us know if there is anything you'd like us to cover.

[@nezztribe5958]

I like your tutorials. You're a very good teacher !

[@playgroundai]

Thanks so much! Means a lot to me 🙌🏼👍🏼

[@osagie2]

I just Love it!!😀

[@Sonusingh-nv2kl]

00:05 Introduction to Canvas series 02:36 Learn how to fine-tune and generate images with different styles using an image editing program 05:02 The image quality and details vary based on the settings. 07:36 Changing the seed and altering specific numbers can generate different variations of an AI-generated image. 10:14 Use plms or ddim for quick generations, and Euler or dpm2 for more detailed images 12:43 Generated images can have more variance and allow for manipulation of smaller details 15:08 Canvas offers various editing options including delete, copy, and download. 17:31 You can manipulate images in many different ways. Crafted by Merlin AI.

[@millionseeds7163]

Absolutely brilliant app and desktop. Fair pricing. Amazing AI. Thanks guys. Always keep it real and you'll be successful.

[@playgroundai]

We appreciate the kind words and support!

[@johnbarbuto5387]

It is almost overwhelming. It is as if we have been launched into a new world. How can we be so lucky to have access to so much! I feel sorry for the old folks who can't keep up (well, older than 76, my personal upper limit for "young!"). Cheers!

[@ivan9663]

Just Amazing!

[@milosblagojevicc]

Amazing video man, thank you

[@noavan1]

I always enjoy watching your tutorials

[@playgroundai]

Happy to hear that! Very kind and thoughtful of you 😊

[@wolfsangeleyes]

Excellent tutorial. Looking forward to more.

[@playgroundai]

Thanks so much for the support! More to come soon. 👍

[@mishivanv]

Thank you very much❤ Will be great to learn how to smoothly integrate specific objects in the same picture style, such as notebook on the table. And switch one object to the other - change clothes, for instance.

[@playgroundai]

We will cover this in a future video to come soon!

[@cekuhnen]

What I find playground missing are layers Each new generation is a new layer This way you can compose and edit individual elements

[@schmutly]

Awesome explainations and details as usual. Thanks and waiting for the next one ^_^

[@playgroundai]

Glad it was helpful and yes more to come soon! 👍

[@AKSenor24]

Great introduction. Many thanks!

[@playgroundai]

You are welcome!

[@amigazo3972]

Thanks a lot for this introduction!

[@playgroundai]

You’re welcome! 👍🏼

[@ManojPrasanna]

Really awesome, and thank you for the explanation.

[@playgroundai]

My pleasure!

[@user-wk5xj5po8u]

Thank you! This was really informative.

[@playgroundai]

You're welcome!

[@ericf2091]

hello, I love Playground, I just found out, after having made some steps on Midjouney, I’m a beginner whis the IA, and really I love Playground, I’m impressed by the images that can be done with it! (Too bad that the hands are currently a problem (difficult for me to retouch them, so I sometimes have to erase beautiful images because of this). But apart from that a big thank you for your fantastic and "magical" work! greetings

[@playgroundai]

Glad to hear you are enjoying playground! As for the hands, there are workarounds, a good model (is always key for good hands) Sometimes you can start with one filter or even just SDXL and generate an image with a good hand and use img2img to change the look and style with filters. You can also use controlnet and a good reference image to get good hands. We are working on a better inpainting model that can help with hands too but the team is still working on it. I will be covering this topic very soon. 👍

[@pedrodelacabra1855]

Excellent. Thx so much. I learned alot 👍

[@playgroundai]

Glad it was helpful! More to come!

[@absinthesesotericadventure7631]

that was really easy to follow, thank you so much!

[@playgroundai]

You're so welcome!

[@richardshepherd9992]

Awesome tutorial, even more awesome application, well done guys.

[@playgroundai]

Thanks so much! Appreciate the positive feedback. 👍

[@marziopessoa]

Another nice work. Thansk for sharing!

[@playgroundai]

My pleasure!

[@toriqussalaam]

WOW! OMG SERIES!! THX 4 THIS!!!

[@playgroundai]

You're welcome! More to come in the following days! Stay tuned!

[@cueball5858]

Excellent!

[@IranForexTraders]

damn, you did great job, i love your work and i will buy the pro plan just to say thank you

[@VehicleInsurance]

So calm voice, almost like ASMR.👌😊👍

[@playgroundai]

😊 thank you

[@diemnguyen6037]

Thanks a lot. Great tutorial. 😊

[@playgroundai]

You're welcome!

[@Greenpie101]

Absolutely loving Playgound AI so far. It has better quality than most imo with all the added benefits of what feels like a legitimate generation suite. THANKS! also the 500/day limit is amazing for a casual user like myself

[@playgroundai]

Glad to hear it and we appreciate your support. Enjoy! 🙌

[@icedmocha6670]

Why cant produce the exact same image using different sampler ? I keep the same seed number, but it always produce different image each time i changed sampler. I use free version.

[@Mr.Artemius]

Total miracle!

[@playgroundai]

🙌 Appreciate it!

[@iliassacharki]

you created something great keep going

[@playgroundai]

Thank you and appreciate the support!

[@MagnusRazer]

Es muy util para Arte y muy principiante para Assets sprites.

[@saborsabroso-ci8lq]

Great Tutorial.

[@playgroundai]

Glad it was helpful!

[@user-wn5zc6iq6k]

it look so easy when u do it in ur vedio i wish i was as good as u but i just started itsso fun to seee the art that can be done with ur homepage

[@playgroundai]

It takes time and you will get there. Time and patience and the willingness to learn. I think people forget this is all still very new so we're all learning together. 👍

[@user-rz6em2xk6r]

... спасибо , мого полезного. 👍

[@craftheart6831]

Thanks ! a lot!

[@playgroundai]

You're welcome!

[@marlonalvez9502]

Good job, very good explanation! Playground AI is the best!

[@playgroundai]

We appreciate it! More to come!

[@I.INFINITI]

Well done! With each new video, AI is improved

[@playgroundai]

Glad you think so!

[@I.INFINITI]

@user-tn8uu2cu8g I have nothing against religion. But this has always been a brake on progress. And why is this so, if not for the reason that only the Lord had to work miracles. For example, a tooth hurts, and only Jesus has to heal it, but now people can do this magic. And when a doctor treats a person with a serious illness, it is practically the resurrection of the deceased. But for some reason, all religions ignore the words that are written in both the Quran and the Bible: God created man in his own image and likeness. And there is nothing wrong with the fact that humanity is progressing, it is a development, it is a gift from Allah. A gift that was given to us from the very beginning.

[@inderus]

Great learning

[@playgroundai]

Thank you! Cheers!

[@MultiBraner]

Amazing

[@playgroundai]

Thank you! Cheers!

[@DavidRisnik]

beautiful and complete explanation of how to use Canvas Playground! What is the usage limit in Free mode? I often get a timeout reached message.

[@theBADsector]

Excellent

[@playgroundai]

Thank you! 🙌🏼

[@mawya_gamer]

Bro this is so cool

[@playgroundai]

Thank you! Hope you enjoy it!

[@mawya_gamer]

@@playgroundai welcome 👍

[@nielsgrote]

great🙂

[@playgroundai]

Thank you!

[@tradehut2782]

I've been using Playground since 2 days. It is very very nice. Kind of switched from Leonardo because in Leo the free plan doesn't give as good images

[@playgroundai]

Welcome and appreciate the support!

[@tradehut2782]

@user-tn8uu2cu8g quiet!

[@a.rainha.da.cocada]

Woowww!!! 👏👏👏👏

[@playgroundai]

cheers!

[@JonathanPerron76]

merci :)

[@playgroundai]

You're welcome!

[@nikolasreissis5603]

Thank you very much for the thorough tutorial. I'd like to report that the copy and or paste option does not show up in Firefox neither it works with ctrl+c and ctrl+v. It does show and work in Crome.

[@personwhohasayoutubechannel3]

Hey been having fun with this awesome site for the past year, but if you could please - it would be nice to have an on-site representation of what everything means.

[@VitorGalveia]

Amazing video, it explains a lot about this application. One thing I tried to do was to transform the raccon in a way that one eye is winking and I couldn't make it. Is there a limitation on this tool to make such thing?

[@playgroundai]

Glad it was helpful. I haven't tried to do a winking photo but you can try describing it in a different way, the AI may not know what winking is. Perhaps one eye closed or you can try doing 2 images and combining one image with the eyes open and the other closed., just delete the areas around it. Or maybe inpaniting can work too. If I have later I'll see what works best.

[@peteskyrunner4845]

Great tutorial. When you drag the purple box it snaps to the edge of the image. How do you turn that on please. I struggle to do it exact when doing manually.

[@playgroundai]

Not sure what you mean? Can you point out where in the video you see it? By default there are alignment tools built in that help with lining things up.

[@irinaivanova8119]

Couldn't find "Storybook" option in filters

[@jonrend]

nice tutorial. I think this is the best flexible AI art tool I have come across. Is this using Stable Diffusion XL (SDXL) version 1.0?

[@playgroundai]

Appreciate it and yes SDXL 1.0 but we do still have filters that use SD1.5.

[@arnaudvaratges1557]

Hello and thank you for your tutorials. When I superimpose two images, I can't find the copy into function with the right click to unite the two images once pasted. Is there a trick? Thank you for your attention.

[@playgroundai]

You have to select both images then copy and paste, either select both by dragging the canvas with your mouse and select both images, then CTRL+C to copy, CTRL+V Paste. Alternatively you can click on both images to select both.

[@arnaudvaratges1557]

Thank you for your prompt reply. But when I copy the images, they remain split. They don't unite.@@playgroundai

[@noctunalvidici8851]

How much does it cost to remove censorship?

[@velvetbees]

Canvas is very powerful. I would love to see an option to recolor an image in image to image with a different color palette. In other words, you can have the same image, but you can change it from something like primary colors to tertiary or monochromatic or triadic coloring.

[@playgroundai]

Thanks for the feedback. That would be nice to have. 👍

[@marcorockabilly3422]

I'm using this tool a lot, but not yet logo to make the outpainting work

[@Otchengazoom]

I've tried to add something to existing image, but unsuccessfully. Is there an option to put a tattoo to the neck of the person at my image considering the curves of the body and how to do that?

[@fictionaddiction4706]

Thanks for the tutorial, I love Playground AI. I hope it gets more acknowledgment as it should. Btw What is the difference between the object eraser and the simple eraser?

[@playgroundai]

Great to hear and we appreciate the support. The simple eraser is more like the traditional eraser. It will delete everything even the background which would require you to regenerate that area. The object eraser will only delete the object that is masked and leave the background.

[@user-hx2vy7sb7g]

In Canvas, How can I connect 2 similar images together, especially the pattern on each image. I want to join them to be one big endless flow pattern. What prompt should I use for inPaint generation frame?

[@lisam6992]

Canvas is amazing

[@playgroundai]

Great to hear!

[@nateyen724]

When i generate 4 photos at once, how can I use seed to reproduce the second image? As i am able to recreate the first image out of a set of 4

[@albertomartinez9662]

Hi. This is great. I want to know if I can upload a portrait picture on myself and convert it into somethin else, like anime or cartoon...?

[@playgroundai]

Yes however it will change how you look when using image to image. It's to preserve the identity of people.

[@burhantohfa8810]

I want to ask that is it free to use canvas images for commercial uses as i m having a free plan?

[@2Heavy]

Excellent video. Still can't believe what can be done with AI, and we are only really at the beginning! Can I just check something, though? When you explained the Publish part it sounded like the checkbox determines whether you want the image to post to your profile as public or private, but the description text reads as though it will make the canvas file public, not just the image itself. Is this correct? Thanks.

[@playgroundai]

Yes that's correct. If you don't check that, it will still publish it publicly and you would have to make it private after. Great question as I should've made that point more clear 👍

[@007HANDMAN]

if necessary, I can collect all the good words that the commenters write below. I am very new to this AI, 1 hour before writing this message, from a quick glance you can see that it is different from other AIs. I congratulate you and wish you to continue with other things differently. But I wanted to ask if there is a tool here or somewhere else that can fix the perspective of an image. Thank you😮‍💨

[@playgroundai]

I'm not sure what you mean? Like from wide angle to close up? Or side view to front view?

[@rossdanielart]

Is there a way to generate images with the same quality as the preview? I actually like the look of the preview for concepting and would like if the generated art looks close to it in terms of the grainy quality. For short I would just like to upscale the preview.

[@playgroundai]

The actual preview? No not at the moment, it's configured differently but you can probably mimic it in the settings, low cfg, low quality and details and a certain sampler.

[@rossdanielart]

@@playgroundai Yeah I tried that, the generations isn't as organic as the preview. Thanks though. For now Im jus resorting to snapshotting the preview.

[@AlexanderBardow]

I have a question exactly on the 14:41 second. (I am trying to follow the video, while doing what is explained there on my side.) As the video says: If you want to combine two images, then you need to highlight both of them, then do the right click and there will come the drop menu with the "Copy" option in it. Indeed, the drop menu in the video gives the following choices: - Bring to Front - Send to Back -Lock Shapes -Copy -Download -Delete Well, from my side, while I am trying to do the same (combine two images after highlighting both of them and doing the right click) in my version of Playground the options of the drop menu are the following: - Bring to Front - Send to Back -Lock Shapes -Download two images -Comment -Delete Thus as a user I have a question: How do I copy the two images and then paste them? What exactly do I need to do?

[@playgroundai]

Playground is optimized for Chrome, so if you use something like Firefox, maybe safari, it could be missing some options. Alternatively you can just use the short cut keys CTRL+C, CTRL+V to copy and paste.

[@user-xh6hf2df8x]

HI! Thanks for this video, its very informative and easy to understand. can you please help me with graphic logo creation. not necessarily pictures but text?

[@playgroundai]

You’re welcome! Have you watched this video yet? How to Create Text Effects with Control Traits in Playground kzbin.info/www/bejne/r4jEaHWietWIn68

[@bluerun22]

@15:04 did anyone else notice when he prompted pixar style background, it brought in a Phallic mountain? Classic disney!

[@joseraulsotoflorez5287]

I just saw a video you make about LeonardoAI and now this Playground, what do you think about both, as I could see this one is better I believe.

[@playgroundai]

That was before I was employed with playground. Of course I would be biased but I agree with you 😁👍

[@poolcreates]

I can listen to this man talk for days, really speaks like Keanu Reeves

[@playgroundai]

😊 you’ve actually made me blush 😬 thank you so much that’s quite the compliment and I’m a huge fan of Keanu!

[@poolcreates]

@user-tn8uu2cu8g shut up goofball

[@jasonblackie7410]

We have blog posts with a certain theme for the main images. Is it possible to use these images to change how they look depending on the blog subject?

[@playgroundai]

I'm sorry I don't really understand what you are asking, can you clarify?

[@jasonblackie7410]

So, we have had a 3rd party create images for each of our websites blogs. These images have a specific theme which I am having difficulty recreating myself for new blogs. Can Playground create new images based on my current blog images?

[@a_vin7644]

With what model by default canvas works. If I can change model?

[@MonzonMedia]

by default it uses SDXL, when you switch to certain filters like realistic vision, rev animated etc it will switch to SD1.5

[@jagadish63]

Its a wonderful presentation, thanks. I found it strange that whenever i asked for a pig and a dog dancing, i kee getting two pigs, never a dog. Why does this happen?

[@jorgefuenmayor2314]

I want to use the draw to edit pencil, but when I push the generate buttom, does not do anything

[@Valent-eh4lt]

Please advise me what to do if I don't have the Copy option in the dropdown menu - at 14:38 - after the phrase "If you drag on a canvas... select...right click...COPY". Thank you in advance.

[@playgroundai]

You can also use control+c to copy and control+v to paste. 👍🏼

[@mashitup4594]

can´t copy neither. @@playgroundai

[@TheLastDispensation]

Does this come with canva I pay for?

[@timedriverable]

Can this match Leonardo's "alchemy" as far as more sophisticated realistic photo or painting details...outside of that a great informative introduction.

[@playgroundai]

I'd say so, with the right prompts and filters it can be achieved.

[@jamesmorrison7876]

I am enjoying using Playgound. What is the difference between Board and Canvas? Is one better than the other? Seems like they are the same except the display. Can u clarify?

[@playgroundai]

Great question James! Board is more for quick generations and also great for mobile users as we don't have a mobile app at the moment. But Canvas has more editing tools like outpainting, object eraser, instruct to edit, draw to edit etc. Think of canvas like your touch up section and board for quick and easy generations. Many people use both, create in board then edit in Canvas. Personally I use canvas 80% of the time now since I work from a desktop. Hope that helps! 👍

[@NeXusAG]

Can i place another on one image? Can i add 2 of my own images for this?

[@Yeomjinhyeok1006]

what the hack!!! Nice

[@playgroundai]

🙌Appreciate it!

[@cekuhnen]

Hey I am curious what your suggestion is for using AI to create vector like logos. It can remain being a bitmap. I am trying to create an assignment where students develop a logo with AI. Claas

[@playgroundai]

My friend, you know more thank you think 👍 Here's an example prompt "text saying "wolves", 2d ferocious blue and white wolf, vector illustration, angry eyes, football team emblem logo, 2d flat, centered." Note that the text won't always come out perfect and may require a few re-rolls and you can't really control the font and size. If you don't need text, just things 2d concepts like vector, 2D, logo, emblem etc. Using something like football team logo or similar also helps influence that style. Hope that helps!

[@cekuhnen]

@@playgroundai duh I do not see the forest for the trees - "2D logo of ..., vector illustration, 2d flat, centered". This works very well. I have to see how this could work also with an image input with SDXL. The Playground result can be traced in Illustrator / Inkscape to clean them up.

[@playgroundai]

It will take some experimentation to see what works for you, SDXL is slightly better with natural language so use the terms you know and think what would a stock image be tagged with. Usually works for me or gets me close to the idea. 👍🏼