I've been working with Photoshop for over two decades now, and I'm still very impressed by how the software continues to develop. No matter whether, new functions and features come directly from Adobe or through third-party extensions like AlpacaAI to expand the range of functions. For me, indispensable tools for expressing my thoughts and ideas in illustrative image montages. My journey started with photography end of the 90s, followed by digital image and graphics editing in the early 00s, and now I'm diving deeper and deeper into these topics as a full-stack developer since 2010. Even though at the end of the day I might only use a fraction of it, but I followed the development of AI with much passion. Thanks for sharing.
Thank you for sharing also! The image and digital art space has hit an exciting and controversial time, I'm really excited to see where it goes as well :)
These PS plugins you're finding and featuring are game changers. The PS landscape is being re-built and re-defined almost overnight. We're living in the greatest period of accelerated learning in the history of Mankind. It's a privilege to be sailing on this ship at this time. Many thanks, as always for your detailed and dedicated approach. ⛵️🌊
Thank you man! Not wrong, it's insane how fast it's changing and photoshops becoming this massive platform beyond d the insane tool it already was. Exciting times :)
Alpaca doesn't work badly, but as a photographer and designer I can say that generative fill still does a much better job. Of course, this will vary depending on your usage area. I used gen fill to fix small details and problems in AI photos like hands or in areas that appear corrupt and it always helped me solve the problems, but I cannot do the same with alpaca. That's why it's great with creative content, but fails in the detailed fill section.
Instead , you can use , Krita , + Krita Ai diffusion plugin ( both are open source ) , You will have the option that is a way better than gen fil , you are free to use any modal, lora etc.. you want , Also you have the live features to sketch or gen in real time .( less than 2 s for my case in my rtx 3070 ) , give it à try
As an artist, this makes me very conflicted. But I would like to try using this to help accelerate my process, so that it may give me references to work off of.
I'm more a casual Photoshop user, but yes, I feel the same about AI: use it with moderation, to help speeding up parts of your work, but without dehumanizing the whole process. Otherwise we just sit and write prompts and the whole work is made by AI. I guess at some point it will be like that, sadly... I believe and I hope that when AI will take place in a stable way in our life (will take over? ehehe), then human beings need a "basic birth income", to enjoy life at least. This is the only way humanity can live with AI substituting a lot of works previously done by humans, without turning the majoirty of people into jobless outcasts of the society.
It's understandable to feel conflicted as the world changes, but there's nothing to truly be conflicted about. No profession or skill or technology or anything else deserves a right to be limited to to those "highly skilled" in it just because greater access might diminish or outright eliminate a field. I used to think the same way, but now I'm grateful that AI and other tools are opening up avenues for more and more people to be creative and I'd rather sift through and excessive mountain of "shit" to find true gems (whether in art, music, etc) than continue to exist in a world where people are limited in the scope of what they can do. I've used AI to design album covers over the last year. I can throw in prompts and get hundreds of ideas in the same time a human artist could provide just one. I can adjust ideas and evolve them into what I'm aiming for, all without the need of paying an arm and a leg for one concept. I get it, it sucks for people who've spent their lives training to be an artist in whatever medium they're in, it sucks to be replaced or feel that it's heading that direction, but no one is irreplaceable and the advancement of technology shouldn't be squashed because people feel entitled to what they do. As AI gets better and better and the need for human artists diminishes, there will still always be a place for them in the same way horse and buggy drivers still fill a niche. AI isn't going anywhere, but whiny self-important artists sure are. Haha.
the StableDifussion depth option is pretty cool to have inside of Photoshop, by saving it as a selection it can be used inside camera blur and get some cool Shallow DOF/bokeh effects. I usually make my "approximate/cheap" masking manually but this is a game changer. I knew Alpaca as a chatGpt competitor but wasn't aware of this plugin (if they're in fact the same people) Thanks a lot for the video, pretty useful info!
Thanks for the insight mate, I like the idea of combining it with blur, I didn't consider that but will need to have a play with it :) Much appreciated!
Tryn to work out the accent, reckon either Melbourne or Sydney. Good on ya mate. BrisVegas here. Subbed. BTW I use SD A1111 PS plugin and getting pretty good returns and streamlined wrkflws. Offers more options for those who work outside the Alpaca ecosystem.
I have some questions...1) Does this plugin require internet access, 2) Is it overly controlled/censored like Dall E, or Midjourney, refusing to do anything "offensive" or copyrighted, and 3) How much does it cost? Can I pay once, or does it require a monthly sub?
No problem: 1 - internet access is required as images seem to be processed on the server 2 - I haven't tried to work past the censors sorry, so not sure Also it is subscription only
thank you for introducing me to the existence of the alpaca! I spent a day to install Automatic1111 locally and compile\install Auto Photoshop SD plugin, but when googling on its docs - found this vid :) With Alpaca I think that local SD and Auto PS SD is unnecesary, hehe ;)
Your video showed up on my yt and I rly enjoyed it, subbed :) Got any good AI tools to recommend for generating stock like images, of people especially, That can be used for marketing purposes or om websites?
Thank you! At this stage I still think Midjourney does the best, I just use Topaz Photo AI or Topaz Gigapixel to upscale. Otherwise, OpenArt.AI is good at times and has a stock photo generator (you can try it for free I think). Its not perfect but it does use an AI tool to fix faces so it could work for you. Otherwise, if they dont do it for you, Leonardo Ai's photoreal feature is good, and Playground AI has a great upscaler which makes images really high resolution for an image generator. It comes down to personal preference from there :)
@@rbbdzz oh yea 😅 Luminar Neo also does it, but I find the masking tools are a bit basic and lack finesse. You can also get a nice subtle depth of field effect in light room by selecting the background and reducing clarity or sharpness (with a linear gradient subtraction if needed). Although that method works better if you’ve already got a bit of out of focus going on already.
Hey mate, it's a real shame about Alpaca :/ However, Leonardo's Realtime Canvas is very similar to their web application, here's an older video of itbefore it was renamed: th-cam.com/video/geYXRPIwQXY/w-d-xo.html
It works well, however for that specific purple, I do think Gen fill is better - but this has so many other features gen fill doesn't. Both are the ultimate combination for Photoshop
that sketch feature looks like a feature they‘re gonna unveil as great adobe sensei innovation at the upcoming MAX. I mean its absolutely amazing and they just have to include that in Phitoshop. Imagine Fresco supporting this. Worlds gonna get even more crazy (or lazy)
@@WadeMcMaster like it always was; I tend to compare it to the industrial revolution; didn't make us all superlazy; some fell victim to exploding boilers, but just like that it might take a decade or two to really land in a way it really makes sense(i) for the rest of us.
Oh! By the way, what are these images trained off of? One of the reasons why I'm less against adobe's generate of AI is because they are using licensed stock. What is alpaca using for learning/training?
@@WadeMcMaster I tried it and it works fine. The only thing was I had to extract the .ccx file with 7-zip. So that was annoying but it ended up working
wow the depth thing so usefull if on video to add layers of smoke which is so hard to control on set when making a film. i guess davinci has that but never tried it for fog yet..
Yeah, i'm sure there are other uses but thats all I can think of at this stage haha. I know it can be used in controlnet to create images with similar layout and depth.
"Learn how Alpaca, the Photoshop AI art plugin that strong functionality that goes beyond Generative Fill! Use cutting-edge AI techniques to enhance your creative efforts and achieve beautiful, personalized outcomes. Discover how Alpaca can improve your Smythos workflow."
HOW my BRO do you keep up and so sharp and 💯✨⚡️💪🏻 on ALL?!? Wow I can barely keep up (I feel for all my web design clients as I keep getting pulled away from all my work for all the daily AI vids and then playing with ha but they all benefit from in the end, rather constant beginnings!)
You save time from ai, and reinvest that saved time I to learning about ai. It's a win! Haha Thank you mate :) I'm glad you like the line up of vids lately!
I did everything as per the instructions but it doesn't work it doesn't generate anything, i did a rough sketch of a house in ps on a layer and asked for scary haunted house with over grown garden, pressed the sketch button, takes ages to process but it just came back with the original sketch i did.
Not working for me. It keeps showing red about my selections. This never happened in your tutorial. How come anytime I make selections it gives one complaint or the other? Any possible solution to this?
Are there any plugins that can do something similar to Photoshop’s native Generative Fill? Alpaca does not do Generative Fill like Photoshop’s generative remove, crop, and fill.
not bad at all but, really frustrating when it comes to work with big images. I mean... "big". I was trying to fill 2 sides of a picture and both were someting like 1350px each and Alpaca can't do anything with 2048+ images. Not bad anyway.
@@WadeMcMaster Yeah. But still... Works better than native generative fill and you don't have to deal with prompt hacks to jump over the Adobe censorship on everything.
Under tools - sketch and then there's 'Sketch settings, whichyou click to expand - it should be there under Sketch Fidelity & Color Layers. I've never encountered it not being there so if it doesn't appear, I would update and then contact their support if it still doesn't work.
Why does this Alpaca AI support normal and square selections? But when a selection is round and changed and not normal, artificial intelligence automatically makes it normal and square How do I solve this problem?
Sorry mate I'm not sure if this will help, but you can use fill, and I believe it works on transparent areas in layers. So you could use your selection, delegate part of a layer and use fill within that area
Sorry mate I'm not sure if this will help, but you can use fill, and I believe it works on transparent areas in layers. So you could use your selection, delegate part of a layer and use fill within that area
Hey mate thanks for the reply? Just to check do we need at least a layer that has a flat color as reference when using sketch? At times my doodle do not have the color out on screen
@@WadeMcMaster No.For example I want to change only one item in the photography.let's say there is a truck and I want to change it to car,or there is a black car,I want to change its color,or in the mountain there is no lake but I want to add a lake,etc.
No worries, just head to their website, sign up, login and download the plugin, then ddouble clickthe install fiel and it'll walk your through. There are some steps here if that helps: docs.alpacaml.com/home/installing
Sorry to hear guys, I never experienced that issue personally - wish I could be more helpful there. I agree though, a plugin which processes locally would be amazing!
The thing with adobe firefly that i find great is that it uses ONLY ADOBE STOCK pictures as a data training set. So with the "vanilla" version, i'm okay... cause no stealing data. What data set does this plugin use for the ai training though? Is it scraped off the internet from unknowing/ unconsenting artists? Because if that's the case I don't want it.
oh ps Adobe are currently working out a settlement as there model has images that where copyrighted. adobe hushed media from having it on the news etc by putting a injunction till the dispute is settled.
Fill is not even close to ps generative fill. You can only make square selection. I was trying to make simple task as add sunglasses and result was really bad. Alpaca changed skin tone, selection edges so you have to retouch... on the other hand is not bad for sketch art.
I definitely think is superior for generating full images, but that was my conclusion too, generative fill was better. I did a direct comparison not long ago
clearly using stable diffusion like adobe is using, anyway of using cititai models with it and does it have gen expanded feature? hopefully it will have some sort of free credits when in full swing being on disability pension cant afford subscriptions, i wish someone would make something like this that runs of our local Stable diffusion installations. but its early days for SD and clearly by adobe and this people are seeing its use and it will spawn many more like it. tried it and it has a hard time working out the base image and using the over all photo as the controlNet image so basically it can't edit expand etc as the selected are is seen as the only information it has. where adobe are clearly using all layers as a snapshot and compression for a controlNet image.
Yeah, Stable Diffussion is growing vastly across platforms like this. Alpaca only lets you train your own model though as far I can see. I agree with you though, it maybe better for the features it does have, while adobes generative fill maybe better for filling and expanding.
Yeah for sure. You can type up the text, control click the layer and give it a texture. Or you can click around text to blend it into something or a combo of both. You just gave me video ideas lol
@@WadeMcMaster Yeah if you could make a video on how to apply AI decoration to text that would be awesome. I don't even know if it's possible but I read in Alpaca's discord that they are working on a solution.
I'm just reading this comment now, and I am actually already working on a video! haha. I'm not super satisfied with the results but some of it will be useful at the least.
That may have been because I signed up earlier I think, I could be wrong. Also the guys at Alpaca may have added that to my account as I am in contact with their team on odd occasion.
I wonder if we are heading for a future where Non Ai artwork will be super rare and valuable because nearly everyone stopped learning how to draw and paint. naaa
![เปิดใจ "อาจารย์เบียร์ คนตื่นธรรม" อดีตเคยสุดทุกทาง ถ้าใช้หนี้หมดไปบวชแน่ l [Nickynachat]](http://i.ytimg.com/vi/IxeE0xVhy08/mqdefault.jpg)
I've been working with Photoshop for over two decades now, and I'm still very impressed by how the software continues to develop. No matter whether, new functions and features come directly from Adobe or through third-party extensions like AlpacaAI to expand the range of functions. For me, indispensable tools for expressing my thoughts and ideas in illustrative image montages. My journey started with photography end of the 90s, followed by digital image and graphics editing in the early 00s, and now I'm diving deeper and deeper into these topics as a full-stack developer since 2010. Even though at the end of the day I might only use a fraction of it, but I followed the development of AI with much passion.
Thanks for sharing.
Thank you for sharing also! The image and digital art space has hit an exciting and controversial time, I'm really excited to see where it goes as well :)
These PS plugins you're finding and featuring are game changers. The PS landscape is being re-built and re-defined almost overnight. We're living in the greatest period of accelerated learning in the history of Mankind. It's a privilege to be sailing on this ship at this time. Many thanks, as always for your detailed and dedicated approach. ⛵️🌊
Thank you man! Not wrong, it's insane how fast it's changing and photoshops becoming this massive platform beyond d the insane tool it already was. Exciting times :)
Alpaca doesn't work badly, but as a photographer and designer I can say that generative fill still does a much better job. Of course, this will vary depending on your usage area. I used gen fill to fix small details and problems in AI photos like hands or in areas that appear corrupt and it always helped me solve the problems, but I cannot do the same with alpaca. That's why it's great with creative content, but fails in the detailed fill section.
I have found gen fill is awesome for photos, less so for digital art. So I agree, it's about differing strengths and what you need it for!
Instead , you can use , Krita , + Krita Ai diffusion plugin ( both are open source ) ,
You will have the option that is a way better than gen fil , you are free to use any modal, lora etc.. you want ,
Also you have the live features to sketch or gen in real time .( less than 2 s for my case in my rtx 3070 ) , give it à try
As an artist, this makes me very conflicted. But I would like to try using this to help accelerate my process, so that it may give me references to work off of.
I feel the same
I'm more a casual Photoshop user, but yes, I feel the same about AI: use it with moderation, to help speeding up parts of your work, but without dehumanizing the whole process. Otherwise we just sit and write prompts and the whole work is made by AI. I guess at some point it will be like that, sadly... I believe and I hope that when AI will take place in a stable way in our life (will take over? ehehe), then human beings need a "basic birth income", to enjoy life at least. This is the only way humanity can live with AI substituting a lot of works previously done by humans, without turning the majoirty of people into jobless outcasts of the society.
It will expand your artist creativity ✨😉
Exactly!
It's understandable to feel conflicted as the world changes, but there's nothing to truly be conflicted about. No profession or skill or technology or anything else deserves a right to be limited to to those "highly skilled" in it just because greater access might diminish or outright eliminate a field. I used to think the same way, but now I'm grateful that AI and other tools are opening up avenues for more and more people to be creative and I'd rather sift through and excessive mountain of "shit" to find true gems (whether in art, music, etc) than continue to exist in a world where people are limited in the scope of what they can do. I've used AI to design album covers over the last year. I can throw in prompts and get hundreds of ideas in the same time a human artist could provide just one. I can adjust ideas and evolve them into what I'm aiming for, all without the need of paying an arm and a leg for one concept. I get it, it sucks for people who've spent their lives training to be an artist in whatever medium they're in, it sucks to be replaced or feel that it's heading that direction, but no one is irreplaceable and the advancement of technology shouldn't be squashed because people feel entitled to what they do. As AI gets better and better and the need for human artists diminishes, there will still always be a place for them in the same way horse and buggy drivers still fill a niche. AI isn't going anywhere, but whiny self-important artists sure are. Haha.
the StableDifussion depth option is pretty cool to have inside of Photoshop, by saving it as a selection it can be used inside camera blur and get some cool Shallow DOF/bokeh effects. I usually make my "approximate/cheap" masking manually but this is a game changer. I knew Alpaca as a chatGpt competitor but wasn't aware of this plugin (if they're in fact the same people) Thanks a lot for the video, pretty useful info!
Thanks for the insight mate, I like the idea of combining it with blur, I didn't consider that but will need to have a play with it :) Much appreciated!
Can you send me the firefly update version?
Tryn to work out the accent, reckon either Melbourne or Sydney. Good on ya mate. BrisVegas here. Subbed. BTW I use SD A1111 PS plugin and getting pretty good returns and streamlined wrkflws. Offers more options for those who work outside the Alpaca ecosystem.
Very cool, I'll check that out! Thank you for the sub and also... I'm just a few hours north of you in Maryborough (30mins from Hervey Bay :) ).
Very cool! For upscaling, I find photoshop's native upsclaer works very well.
It's not bad, I generally use Topaz Gigapixel these days, but they all have their place too!
I have some questions...1) Does this plugin require internet access, 2) Is it overly controlled/censored like Dall E, or Midjourney, refusing to do anything "offensive" or copyrighted, and 3) How much does it cost? Can I pay once, or does it require a monthly sub?
No problem:
1 - internet access is required as images seem to be processed on the server
2 - I haven't tried to work past the censors sorry, so not sure
Also it is subscription only
thank you for introducing me to the existence of the alpaca! I spent a day to install Automatic1111 locally and compile\install Auto Photoshop SD plugin, but when googling on its docs - found this vid :) With Alpaca I think that local SD and Auto PS SD is unnecesary, hehe ;)
Cheers mate! Glad I could help :)
Man this is straight fire! Thank you so much for sharing!
No problem, thank you!
Your video showed up on my yt and I rly enjoyed it, subbed :)
Got any good AI tools to recommend for generating stock like images, of people especially, That can be used for marketing purposes or om websites?
Thank you! At this stage I still think Midjourney does the best, I just use Topaz Photo AI or Topaz Gigapixel to upscale. Otherwise, OpenArt.AI is good at times and has a stock photo generator (you can try it for free I think). Its not perfect but it does use an AI tool to fix faces so it could work for you.
Otherwise, if they dont do it for you, Leonardo Ai's photoreal feature is good, and Playground AI has a great upscaler which makes images really high resolution for an image generator.
It comes down to personal preference from there :)
dude, you can use depth maps direcrtly in the lens blur to make a super accurate and real looking depth blur, you should have shown that xD
Yes I should have lol.
Thanks. Works OK in Photoshop 25.0, installing *.ccx fail but if copy alpaca folder to C:\Program Files\Adobe\Adobe Photoshop 2022\Plug-ins - is OK
Good to know! Thank you!
That depth mask would be great for creating a shallower depth of field with a blur filter
That's a cool idea!
there is native thing in Photoshop neural filters called Depth Blur, just try it
@@rbbdzz oh yea 😅 Luminar Neo also does it, but I find the masking tools are a bit basic and lack finesse.
You can also get a nice subtle depth of field effect in light room by selecting the background and reducing clarity or sharpness (with a linear gradient subtraction if needed). Although that method works better if you’ve already got a bit of out of focus going on already.
True, it works well too
Alpaca c'est fini ce 25 octobre zut de zut, je viens de recevoir un mail. Connaissez-vous un programme similaire ...
Hey mate, it's a real shame about Alpaca :/ However, Leonardo's Realtime Canvas is very similar to their web application, here's an older video of itbefore it was renamed: th-cam.com/video/geYXRPIwQXY/w-d-xo.html
depth map can be usefull in filter - blur - lens blur for adding controlable blur like depth of field effect
Thats a cool idea!
How does it work for adding elements into an existing image? Gen fill does a good job but could be better.
It works well, however for that specific purple, I do think Gen fill is better - but this has so many other features gen fill doesn't. Both are the ultimate combination for Photoshop
Thanks a lot brother. ❤❤
that sketch feature looks like a feature they‘re gonna unveil as great adobe sensei innovation at the upcoming MAX. I mean its absolutely amazing and they just have to include that in Phitoshop. Imagine Fresco supporting this. Worlds gonna get even more crazy (or lazy)
Yeah exciting times hey. Many will get lazy, but a few will go next level and will create some awesome stuff!
@@WadeMcMaster like it always was; I tend to compare it to the industrial revolution; didn't make us all superlazy; some fell victim to exploding boilers, but just like that it might take a decade or two to really land in a way it really makes sense(i) for the rest of us.
Great video! Subscribed!
Thanks for the sub! Have a great day :)
Oh! By the way, what are these images trained off of? One of the reasons why I'm less against adobe's generate of AI is because they are using licensed stock. What is alpaca using for learning/training?
I hear you, it's a big time saver. Om not sure where their dataset originates from sorry.
I hate to be that guy but, does it work with crack photoshop (v25.0.3 or 25.3)😅😅
I don't know sorry mate, I haven't tried.
@@WadeMcMaster I tried it and it works fine. The only thing was I had to extract the .ccx file with 7-zip. So that was annoying but it ended up working
in which version of photoshop this plugin is supportive? can i use it in adobe 2019 version?
it appears the earliest compatible is Adobe Photoshop 2021, at least v22.5.9
wow
the depth thing so usefull if on video to add layers of smoke which is so hard to control on set when making a film. i guess davinci has that but never tried it for fog yet..
Yeah, i'm sure there are other uses but thats all I can think of at this stage haha. I know it can be used in controlnet to create images with similar layout and depth.
"Learn how Alpaca, the Photoshop AI art plugin that strong functionality that goes beyond Generative Fill! Use cutting-edge AI techniques to enhance your creative efforts and achieve beautiful, personalized outcomes. Discover how Alpaca can improve your Smythos workflow."
HOW my BRO do you keep up and so sharp and 💯✨⚡️💪🏻 on ALL?!? Wow I can barely keep up (I feel for all my web design clients as I keep getting pulled away from all my work for all the daily AI vids and then playing with ha but they all benefit from in the end, rather constant beginnings!)
You save time from ai, and reinvest that saved time I to learning about ai. It's a win! Haha
Thank you mate :) I'm glad you like the line up of vids lately!
Sick video!!
Thank you man! I just popped one of your songs into a video for next week ;)
I did everything as per the instructions but it doesn't work it doesn't generate anything, i did a rough sketch of a house in ps on a layer and asked for scary haunted house with over grown garden, pressed the sketch button, takes ages to process but it just came back with the original sketch i did.
ok now it worked
I'm glad it worked mate!
Does it require dedicated graphics card?
I think it's done on their server, not locally.
Not working for me. It keeps showing red about my selections. This never happened in your tutorial. How come anytime I make selections it gives one complaint or the other? Any possible solution to this?
Try a smaller selection or using a smaller image maybe? I think the max window size is 1080x1080px so it may be the area you're working on is too big
@@WadeMcMaster okay thanks, bro.
It does not work on lasso selections. It must be using the rectangle selection tool . youre welcome.
Are there any plugins that can do something similar to Photoshop’s native Generative Fill? Alpaca does not do Generative Fill like Photoshop’s generative remove, crop, and fill.
I don't think so, however you could try alkaidvisions plugin? I don't think it's quite as good but worth checking out
which photoshop version? does it work on photoshop 2023?
Yeah I use it ps 2023
How do you expand an image like GEN FILL. It just select the whole canva and change everything
It would be a matter of expanding the canvas and using the fill feature.
What about the FILL and TRANSFER options ???
They're handy, not bad. I actually recorded them but they kinda slowed the video down.
Is it possible to use for commercial purposes.?
I just tweeted them and all good for commercial use.
not bad at all but, really frustrating when it comes to work with big images. I mean... "big". I was trying to fill 2 sides of a picture and both were someting like 1350px each and Alpaca can't do anything with 2048+ images. Not bad anyway.
Yeah, I look forward to the day Ai handles larger images, they're all a bit limited that way at the moment.
@@WadeMcMaster Yeah. But still... Works better than native generative fill and you don't have to deal with prompt hacks to jump over the Adobe censorship on everything.
I agree mate :)
Is there any sense of what this is going to cost?
Not sure sorry mate. Most of these platforms have tiers from $10, $30 and onward, so it could be similar? I'm guessing though
if i generate the photo with a model, is it ok to use it as a commercial photo?
Yeah, I just asked them and all good for commercial use :)
does it work for the *legal* photoshop version?
That's one I'm using in the video. Latest creative cloud
@@WadeMcMaster I meant for the (il)legal version
Can you fill chosen area only like in generative fill?
You can but (from memory) I believe you need to delete that part of the layer so it's transparent.
I don't see sketch strength, color strength on my mine. Can someone help me. Do I need to update?
Under tools - sketch and then there's 'Sketch settings, whichyou click to expand - it should be there under Sketch Fidelity & Color Layers. I've never encountered it not being there so if it doesn't appear, I would update and then contact their support if it still doesn't work.
Why does this Alpaca AI support normal and square selections? But when a selection is round and changed and not normal, artificial intelligence automatically makes it normal and square
How do I solve this problem?
Sorry mate I'm not sure if this will help, but you can use fill, and I believe it works on transparent areas in layers. So you could use your selection, delegate part of a layer and use fill within that area
Sorry mate I'm not sure if this will help, but you can use fill, and I believe it works on transparent areas in layers. So you could use your selection, delegate part of a layer and use fill within that area
any idea why at times, in sketch mode, the output return are just blank screen?
No, sorry I haven't had that issue personally. But maybe try adding a new blank layer or maybe even slightly changing the prompt
Hey mate thanks for the reply? Just to check do we need at least a layer that has a flat color as reference when using sketch? At times my doodle do not have the color out on screen
🔥
so this will only work on the latest PS version ? ... no crack version ? ... just signed up and Alpaca is compatible with Photoshop 2021 and onwards.
I don't know about cracked versions sorry, I'm on creative cloud
i tried this plugin for anime images, the results were really weird tho, idk its the setting or alpaca just doesnt work for anime thing.
That's fair, I didn't try it for anime actually, so good to know. I found it quite good for more general stuff and photos.
I understood that it does not work with partial selections,as removing object,changing objects etc. like PS.
you mean like semi-transparent selections?
@@WadeMcMaster No.For example I want to change only one item in the photography.let's say there is a truck and I want to change it to car,or there is a black car,I want to change its color,or in the mountain there is no lake but I want to add a lake,etc.
I see what you mean. You do have to make that area transparent, so erase that area. Which is an extra step compared to genfill
Sorry, how do I install the plugin?.thanks
No worries, just head to their website, sign up, login and download the plugin, then ddouble clickthe install fiel and it'll walk your through. There are some steps here if that helps: docs.alpacaml.com/home/installing
t
hanks
what resolution does it work best at?
1024 x 1024 typically
"waiting to start" forever. Doesn't seem to work. hope there will be a local version soon
same problem here...
Sorry to hear guys, I never experienced that issue personally - wish I could be more helpful there.
I agree though, a plugin which processes locally would be amazing!
The thing with adobe firefly that i find great is that it uses ONLY ADOBE STOCK pictures as a data training set. So with the "vanilla" version, i'm okay... cause no stealing data.
What data set does this plugin use for the ai training though? Is it scraped off the internet from unknowing/ unconsenting artists? Because if that's the case I don't want it.
oh ps Adobe are currently working out a settlement as there model has images that where copyrighted. adobe hushed media from having it on the news etc by putting a injunction till the dispute is settled.
I don't know what dataset they've used unfortunately.
does this work with Affinity photo?
I don't think so at this stage sorry mate
is it free or paid?
Currently free, but will be paid eventuallu
Fill is not even close to ps generative fill. You can only make square selection. I was trying to make simple task as add sunglasses and result was really bad. Alpaca changed skin tone, selection edges so you have to retouch... on the other hand is not bad for sketch art.
I definitely think is superior for generating full images, but that was my conclusion too, generative fill was better. I did a direct comparison not long ago
clearly using stable diffusion like adobe is using, anyway of using cititai models with it and does it have gen expanded feature?
hopefully it will have some sort of free credits when in full swing being on disability pension cant afford subscriptions, i wish someone would make something like this that runs of our local Stable diffusion installations. but its early days for SD and clearly by adobe and this people are seeing its use and it will spawn many more like it.
tried it and it has a hard time working out the base image and using the over all photo as the controlNet image so basically it can't edit expand etc as the selected are is seen as the only information it has. where adobe are clearly using all layers as a snapshot and compression for a controlNet image.
Yeah, Stable Diffussion is growing vastly across platforms like this. Alpaca only lets you train your own model though as far I can see. I agree with you though, it maybe better for the features it does have, while adobes generative fill maybe better for filling and expanding.
Not working for me, just sits there, saying 'loading'
That sucks mate, try contacting their support
Подожди подольше, оно генерирует долго в зависимости от конфигурации компьютера, проверял на нескольких pc
Does anyone know a way to do a generative fill on just some text to give it some cool effects?
Yeah for sure. You can type up the text, control click the layer and give it a texture. Or you can click around text to blend it into something or a combo of both.
You just gave me video ideas lol
@@WadeMcMaster Yeah if you could make a video on how to apply AI decoration to text that would be awesome. I don't even know if it's possible but I read in Alpaca's discord that they are working on a solution.
I'm just reading this comment now, and I am actually already working on a video! haha. I'm not super satisfied with the results but some of it will be useful at the least.
is it free? or not?
It was, now its premium but there is a free try trial.
Mine just keeps saying waiting to start.
Ah man that sucks. Try hitting up their support
How do you get the unlimited ♾ credits? Mine started with 9999 credits and it feels like it's not enough.
That may have been because I signed up earlier I think, I could be wrong. Also the guys at Alpaca may have added that to my account as I am in contact with their team on odd occasion.
I wonder if we are heading for a future where Non Ai artwork will be super rare and valuable because nearly everyone stopped learning how to draw and paint. naaa
Definitely not.
Sir, is this totally free
It currently is (last I checked, recently) - but it will bea paid plugin evenutally.
@@WadeMcMaster thanks
Does this have a function to do the Gen AI on selections like the PS one? (ie, changing clothes on a photo of a model)
Yes it does, a tool called fill
how? can you please elaborate a bit about it. I'm trying from few days but i want it to change my clothes , hairstyle and all.@@WadeMcMaster
is it free?
Currently it is, but wont be forever
@@WadeMcMaster okay thanks
👋
i cant use it on beta app... sad
Not sure why that is, I recorded a chunk of the video using beta and it worked
@@WadeMcMaster mean beta Photoshop 😓
@@gorshocheg I'm using it on the latest Photoshop beta today and its working. I'm using it along side of the PS AI Gen stuff too.
Awesome, combining the two would be a good move i think
im using my homies photoshop and i can't install the plug in photoshop SUCKS
That's no good mate, is it a recent version?
It says not compatible even though I have photoshop latest version 25, why?
That is really odd. I'd contact Alpaca's support, that doesnt sound right
@@WadeMcMaster im using the beta photoshop could that be the reason?
I got it to work in beta easily enough. I downloaded the creative cloud file if that helps
@@WadeMcMaster mine however is cracked latest version beta
@@EvokAi did you try before you buy ? or do you have to pay for it before you try and how much is it ?
Is the resolution high or is it low like generative fill
basically the same
And if you don't have paid version of Photoshop ?
No idea sorry man!
I have been using it on unpaid version but now Alpaca introduced credits, so everything falls into the water...
I hear you, unfortunately it all costs money to run so the party always ends at some point.
Does this plugin work with photoshop cs6 ?
I think just PS 2023 and onward
is it free or paid?
currently its free, but will be paid eventually