Stable Diffusion Basic Prompting Tutorial Using PlaygroundAI.com

[@MonzonMedia]

Hi Everyone! Please note that there have been some changes to the pricing format for PlaygroundAI. check it out here playgroundai.com/pricing. Are any of you running Stable Diffusion locally?

[@KiZzymesl]

Yes I run it locally. Thanks for the video! Ran out of usage for Dall-E, but overall the website Playgroundai is a very nice site.🙃

[@KiZzymesl]

So far my fav two local installations are: A111 and CMDR.

[@MonzonMedia]

@@KiZzymesl And thank you for checking out the video! As much as I like Dall-E 2, it's pretty pricey but I like that playgroundai is offering it at a decent price. Appreciate the feedback and support, more to come soon!

[@MonzonMedia]

@@KiZzymesl Great! Yes I like A111 and been meaning to try CMDR. The only other local one I've tried is GRisk GUI but I've settled on A111. I love that out painting is built in to it! May I ask what you are using text to image for?

[@KiZzymesl]

@@MonzonMedia Experimenting with the ideas in the Virtual Worlds. Finding materials and textures at all the right angles seemed almost impossible until now. ☺

[@giri.goyo_yt]

Great job showing the basics. Very methodic and articulate while still having a beginner's pace. No frills, no gimmicky humor. Made me subscribe.

[@MonzonMedia]

@theogore I appreciate that! I always try to get to the point as I know peoples time is precious. I usually end up cutting out a lot of extra footage. 😆

[@gingercooklive]

I am just leaning how to use this. I find your tutorials very helpful, You make it easy to understand.

[@MonzonMedia]

I’m a simple guy and I like simple instructions 😊👍🏼 I’m so glad to hear it was helpful. I’ve got more content on stable diffusion coming soon! Appreciate the support.

[@TheTechBear]

Another great video! Thank you so very much! I learned so much from your first video, and then you told me about this one…I guess I won’t be sleeping tonight! Thank you!

[@MonzonMedia]

Thanks so much, that's so nice of you to say! I'm literally laughing as I type this because I should be going to bed now but I'm still testing some new concepts I'm working on for the next video. 😁 hahaha! Enjoy and more to come soon!

[@DrDaab]

Wonderful. I can't believe how after 3 months of playing w SD I did not get the idea of how you refine the image by moving from txt2img to img2image. So, now I understand a lot more. Thanks !

[@Luxury-Royale]

This is a great video. Very informative. And, I like your calm voice. Also, I have subscribed to your channel.

[@MonzonMedia]

Thanks so much and I appreciate the support! I'm glad it was helpful! 👍

[@twotwinningcavs]

New subscriber from Australia. I’m new to Ai, and art in general and your videos are simplistic enough for me to understand and your definition pop ups are also very helpful!

[@MonzonMedia]

So glad to hear that! Thank you for taking the time to watch and leave feedback. Make sure to check out the other videos.

[@Gamesoldier68]

Thank you very much for the very detailed and instructional video on Playground. I just started using this and I see all of the things I was doing wrong. I hope to have more success in the future thank to your video! Thanks again!

[@MonzonMedia]

You're very welcome! I'm glad it was helpful for you, It's still very new so we're all learning together. 👍

[@youriwatson]

Thanks for the tutorial!

[@MonzonMedia]

You are very welcome! Appreciate the support! 👍🏼

[@angrykanoHoK]

Please make more tutorial about prompts please.

[@MonzonMedia]

Absolutely Angelo, there is so much to cover on it, one video is not enough. More to come soon! Thank you for supporting my channel. 👍

[@MooneLightEntertainment]

This video was really helpful, even though the interface is no longer the same.

[@MonzonMedia]

So glad to hear that Sommy! Yes this was a few months back when playground was fairly new. I'll most likely do an updated version soon. There are a few more on PG on my channel but I also do videos for Playground AI's official channel too. Check it out here kzbin.info/www/bejne/mpy4m3uGhpmhfqM I cover similar concepts except for portraits and the interface is more recent.👍 Appreciate the support and feedback!

[@darrellpidgeon6440]

Perfect content. Playground sounds very interesting. But before signing up on the site, I intend learning more so I don't go in "blind". Sub!

[@MonzonMedia]

No problem, thanks for checking it out. I have a playlist of videos on Playground that you can check out. kzbin.info/www/bejne/npPWYaykibmVnJY

[@giri.goyo_yt]

Can't wait for the inpainting demo!

[@MonzonMedia]

Hoping to get that one edited soon! 👍🏼 I appreciate the support! 🙌🏼

[@ryanhowell4492]

Beautiful

[@MonzonMedia]

Appreciate it! 🙌🏼

[@ryanhowell4492]

@@MonzonMedia I'm still learning

[@jmkmusicpedals]

Really learning a lot from these videos! Does Easy Diffusion have a way to control the Input Strength like you are showing in this video? Is that something I can just put in the prompt? Thanks

[@MonzonMedia]

Glad to hear it although it's a pretty old video, Playground has changed quite a bit. I do most of my playground content on their official channel now. www.youtube.com/@playgroundai As for easy diffusion yes if you are talking about the image strength when using image to image? When you generate an image and hover over it there is an option "use as input" that will trigger the img2img the same way. Then in the settings you will see an option "Prompt Strength" but for ED its opposite, the higher it is, the more variation you get. I've been meaning to do a video on this for some time. I'm over due for an Easy Diffusion video.

[@vanderleymoreirademelo3398]

Desde, São Paulo, Brasil...Obrigado!

[@MonzonMedia]

Hello from Canada and thank you for watching! Brazil wow!!!🙌😁

[@Spongeworthyxxx]

Great tutorial. Do you have any tricks that keep the main subjects from getting cropped? For example your image's wings get cut off at the top. I have tried everything and no luck. Prompts of guitars, cars, people, ... etc keep getting cropped. I would like space all around the subject matter so nothing gets chopped off. Also any tricks to get a completely white background? Any insight would be appreciated.

[@MonzonMedia]

Hey there, well it's expected with ai generated images and it will happen. You can try using negative prompts "remove from image" option below the prompt box. You can put in things like, cropped, out of frame etc and in your prompt add words like centered, symmetrical. It's not going prevent it but it will help minimize it. Another thing you can put in your prompt is describe what the subject is doing, for example if it's a person you can put "standing on the ground or sitting on a chair" to get the full subject in frame. But know the AI is not perfect so you will generate images that are not perfect. Hope that helps!

[@Spongeworthyxxx]

@@MonzonMedia great thank you for the response. I will try those suggestions. Yes AI in this form is pretty new. I give it another six months and things will get even better.

[@angelyukitay6517]

hi i cant see a masking tool or edit background or pose tool in the free playground AI. What I would like to do is using the same characters, I wanted to have different backgrounds and positions. How do I do it?

[@MonzonMedia]

To mask or inpaint you have to select it from actions>inpaint.

[@angelyukitay6517]

May I know how to generate command for the same character for different scenes eg different position and background

[@MonzonMedia]

I have a video I'm working on for this topic. It's a bit tricky and to be 100% transparent, 100% consistent characters is not really possible however there are some workarounds. Stay tuned!

[@svergurd3873]

0:35 "Euler" is pronounced as "oiler". It is German, of course, and after the great Swiss mathematician Leonard Euler.

[@MonzonMedia]

Hi Svergurd, yes I learned that shortly after I made this video 👍 But I didn't know the meaning behind it so thanks for that! I learned something today. 😁

[@svergurd3873]

@@MonzonMedia 👍

[@SuperGari2]

is there to get different poses with same character?

[@MonzonMedia]

Hi Gaby, you can use seeds as described here on the official playground ai channel. kzbin.info/www/bejne/jIu2Z4aKZ6d3gdk You can also do variations and img2img however it will not be exactly consistent, it's just the nature of text to image ai at the moment. Using seeds is a good way though to keep it somewhat consistent. Hope that helps!

[@draeco71]

I have been using playground AI. Any reason why a generated image has multiple faces? Any way we can prevent that through a negative prompt? I have used a negative prompt of multiple images and a regular prompt of single image

[@MonzonMedia]

Hey Mark, it's very common with AI image generation because the data sets are usually trained with a 1:1 aspect ratio, so chances of getting multiple limbs are more likely with other aspect ratios. You can use the negative prompts section and add words like "double head, multiple limbs" etc. This won't prevent it but it will help minimize it. Here's a link to some common ones I use for people tinyurl.com/y4n2ntpj. You can also experiment with sampling methods, Euler, DPM2 etc in the advanced options, some samplers are more prone to duplication but it will also change the look slightly. Appreciate the feedback.

[@robertinavracaric7378]

Hi thanx for video, but my images are deformed :(((

[@MonzonMedia]

Hi Robertina, it's very common with AI image generation because the data sets are usually trained with a 1:1 aspect ratio, so chances of getting deformities are more likely with other aspect ratios. You can use the negative prompts section (remove from image) and add words like "double head, multiple limbs" etc. This won't prevent it but it will help minimize it. Here's a link to some common ones I use for people tinyurl.com/y4n2ntpj. You can also experiment with sampling methods, Euler, DPM2 etc in the advanced options, some samplers are more prone to duplication but it will also change the look slightly. Appreciate the feedback.

[@robertinavracaric7378]

@@MonzonMedia thanks for the quick reply i will experiment...

[@_nanaaaaaa]

Can you make a video how to create a esport logo/company logo using playground ai? I really appreciate it if you make one, tq :)

[@MonzonMedia]

Hi Nasr, that's not a bad idea actually. The logo on my channel was made through playgroundai. Well I mean I had an original black and white logo but I wanted to give it a metallic look and after a few generations I like this version that I currently have. I'll definitely add it to my list of topics under playgroundai as I have a few video ideas. Appreciate the feedback! 🙌👍

[@michaeljp9605]

Playgroundai or Midjourney ?

[@MonzonMedia]

We’ll I’m biased since I work for playground now 😬😊 but in terms of value and versatility, playground hands down. MJ’s image quality is top tier and easy to prompt but you can get the same results with any stable diffusion platform with the right model and prompts. 👍🏼

[@0x444-f7w]

how coud i choose the sutible filiter ?

[@MonzonMedia]

Best to use a filter that fits the style you are going for. Since this video there has been more added to Playground. For example if you want to do portraits, analog, polaroid and instaport are good ones to try. The best way is to experiment!

[@0x444-f7w]

@@MonzonMedia thanks alot , i realy love your tutorials for playground ai it helped me to improve my works so much 💖

[@MonzonMedia]

I'm so glad to hear it! Be sure to check out there other ones on my channel as this is a bit old now. Also, I've recently started working for playground ai so make sure to check out the videos on the official channel www.youtube.com/@playgroundai. You will see my familiar face there too! hahaha! Have a great weekend and thank you so much for the kind words. ☺