Attempted using Leonardo for the first time today and I am really struggling. I use extremely detailed prompts and get nothing close to what I’m asking for. I think this video, after a couple views, may really help me. Thanks for taking the time.
@@WadeMcMaster I really like Leonardo and i feel its easier than Midjourney however, when i try to create things myself it wont let me, says This feature is disabled and cannot use. I was trying to created a simple Welcome Mat with Dog and paw prints and couldnt do it.. Could you do a video on how to create our own stuff without copy and paste someone else ideas? thanks so much
i subscribed to midjounrey a few months ago as it helps with thumbnails for youtube but after watching things for leonardo im very much considering the swap, the characters are cleaner and i love the effect of being able to remove and add backgrounds and add extra bits of the image to expland the image.. some cool features for sure.
I hear you, they are both so powerful. I still prefer Midjourney as I love to prompt on my phone, but its personal preference at this stage as Leonardo is amazing too
If you go through all the community feed images which shows head to toe of human characters, most of them used DreamShaper v5, rarely RPG 4.0, DreamShape 3.2, Deliberate 1.1 but you won't find any other model that successfully rendered full body images from head to toe.
I switched over for a month from Midjourney and honestly, I havent been able to get it to generate a damn thing I could actually use. It seems like a powerful tool but there is also a much larger learning curve that one will have to invest time into before they can fully utilize the program. Time I dont have, and with 15 extra dollars I can access Midjourney, whos prompts seem to have an easier time generating usable content.
I've been using Leonardo to upscale my MJ images with excellent results... but it's a pain using two programs... MJ needs to improve the resolution and price structuring ASAP. Great walkthrough... learned a great deal!
Right. MJ is cuts the butter but they have very limited payment options and don't have their own site/platform like Leo. Leonardo also yields higher res images and provides more credits.
It's pretty crazy! I've been a graphic designer for 20 years and it sped up so much of what i do - not designing but creating smaller elements, its incredible.
Hi, may I say you are a wild and crazy guy!! Great video. I'm new to Leonardo AI, and your video was chock full of stuff I did not know. I can't wait to go try that stuff out. Thanks for your lessons. Super job. May I subscribe?/ OK, Thank you. C'ya.
Image to image and canvas? Image to image will take an image or two, and itnwill use it to guide the creation of a new image. Canvas edits images using ai. Image to image: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE--HSrq1I361A.html Esit using canvas: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-W3Yflfd5u4Q.html
Am I the only one with this problem but I want to generate an image but it says syntax error so nothing comes out and it says I’ve wasted tokens when it hasn’t generated anything
I have heard that they are having some issues at the moment that they are trying to fix. It could be that, they mentioned it on their discord a few days ago. Hopefully it gets fixed soon.
Hi there fellow Aussie! I’ve been playing around with Leo for the last few days & LOVE IT! Would love more videos. I’m not very tech savvy, normally an ink & paint regular artist trying to get same watercolour effects with ai & my iPad. I subbed, cause you are an Aussie! 🇦🇺❤️
I love the site but I have a problem I was doing my little things for a long but after a few days I noticed that the vultures appeared sadly.. I noticed that some of my older creations got more likes I don't want to see any likes on my private drawings by others, these are my works and I am not able to share them with anyone for the moment, I have deleted all the pictures that have liked any idea how to keep away their prying vultures? Or I need to deleted all now?
You can with a paid plan. I use free and am not sure how to do it personally. I found this in their FAQ: Will my content be kept private or can it be used by Leonardo.AI or other users? As per Section 3b of our terms of service, any images that are not toggled to private (or for the duration that they are set to public), grant Leonardo.AI the right to use, copy, reproduce, process, adapt, modify, publish, transmit, create derivative works of, publicly display, publicly perform, sublicense, and distribute text prompts and images you input into the services, or any assets produced by the service at your direction. This license also allows Leonardo.AI to make the assets available to the public and use them in order to provide, maintain, promote, and improve the services. However, if you toggle your images to private, they will not be visible to other users or Leonardo.AI, and they will not be subject to use by Leonardo.AI for advertising or improvement purposes.
I don't know man I could have swore if I looked long enough at that blue rock wall I could see faces. They were more readily apparent when you moved the image around.
I wonder if its possible to organize your images in Leonardo in folders or something like that? It will become chaotic rather fast as your image portdolio grows.
@@WadeMcMaster seriously Wade! AI been around for decades and the last 3 months has been an absolute inflection point (for good and bad obviously for many) but I’ve been trying to share w friends and fam ever since ChatGPT came out that those that don’t embrace will absolutely ‘get left behind’ I still haven’t jumped into Leonardo (but MidJourney daily for sure, in fact today I created all of the Photoshoot pro level realistic images for a health food company company rebrand and they were blown away by the content I created for their new site in 2 hours) The Leonardo Frame fill feature is NEXT LEVEL! Thanks bro for your vids and awesomeness for one and all
Sorry to hear mate :/ I think they've had a huge influx of people signing up since Midjourney removed their free trial so they may have trouble keeping up
Hey man can you tell me how these AI thing works? I got Invoke AI Installed on my computer, it works as some kind of a launcher that can load Stable diffusion models, now the question I have for leonardo is just what the heck is it? Is it a stable diffusion model? or is it a launcher like InvokeAI? I noticed leonardo is categorized as a Sampler, but what does that means? can I download the model leonardo uses as ckpt file and then load it in My InvokeAI? sorry for the questions man but pls help me out if you can
All good man :) I dont think you can download Leonard AIs files (apart from art you create) as its designed as a platform younuse entirely on their website. Leonardo AI does use stable diffussion though, ut is entirely separate from invoke ai. I actually havent use invoke myself.
There is an image area you can include and image prompt! I did a quick video on their image features: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE--HSrq1I361A.html
Hi there. Can you please tell me how to change the color of something, say a dress, while keeping same model??? Thanks. I ve tried masking and erasing, but it always draws back something different to what I expect.
It will generally redraw when using Canvas, I don't think there's a simple recolor mode. I would sya you'd need a photo editor to do it exactly. It's easy in photoshop - ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-S76ZP9ubjYY.htmlsi=m2O-NuOW8-QmByRp Or try photopea, very similar: www.photopea.com/
Have you recieved an email from them approving your account? It can take time. Otherwise i just click launch app and log in. Notnsure whats happening sorry mate
But Leonardo is such a pain in as* to register ,I sign up with them but I am still not allow to use their discord and I have to wait whitelist or something
Hi there, I hope you might be able to answer a burning question I have? I'm totally amazed by this new step in creating art and appreciative of those such as yourself taking the time to help the rest of us who aren't skilled in using this software. I have been wanting to write up some family histories and my own fiction illustrated in the form of graphic novels/manga and not being artistic had thought of commissioning an artist to do that for me. But with the advent of things like Leonardo and Midjourney, the task seems easier. Backgrounds are one thing, but characters are the main objective, either creating an entirely new person or basing it on real people (relatives). Sorry for that exposition... my question is this. Is there a way or some graphics software, etc that can use a generated character and replicate them for different scenes in the story as each iteration of a character through the AI will always be different regardless of how specific you define them? Is there somewhere you might point me to? I did have Manga Studio, but my drawing skills never progressed. I've just found your channel and spending a lot of time going through your back catalogue. Cheers.
Thank you Andrew! I would say the best thing would be to try and create images of a certain character and when you get it, use that image in some image prompts (upload the generated character and keep the same description in your prompt also) and youll be more likely to create a consistent character but still get a few misses. Another step you can take is on w you have a handful of inages that are similar enough, you can train a model in leonardo ai for that character. You could also do what i did by finding a person on pexels or unsplash to reference. Heres how you can train a model: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-EJd54X_T8RA.html I hope that helps! There isnt really a precise way but thats how i would go about it! Thank you again mate, i appreciate you checking out my channel :)
It has a lot mIdjourney doesn't have. It has great quality images (higher resolution upscales too) but not quite as good in variety as Midjourney. It's free to try, I recommend heading over and requesting access so you can give it a go
Great video! Just out of curiosity, what did you base your decision on at 16:45 in the video, when you switch to Stable Diffusion 2.1 - what made you decide to switch to that one for the human face” part of the editing on the image of the arm/hand with the eye? Thanks! 🙏
@@WadeMcMaster I see! Thanks for the reply! Would it be alright if I ask you another question? Today I’ve run into a curious problem with the mask tool myself, and I’m actually a little stumped 🤔
@@WadeMcMaster Now I’m sorry-I didn’t see your reply either, but I definitely appreciate your taking the time to circle back! I can’t entirely remember my question now lol 🤦♂️😅 If I can dig it out of my brain I’ll let you know, and, thank you again 🙏
I'm not sure what's happening there sorry. Leonardo does play up from time to time but could you try resaving the image in a other format? Png, jpg, tiff, etc
i have been tryng to generate a man a a woman with specific physical characteristics, and i have upload some pictures of famous men and women, for the IA to use as reference, but is always something bad, sometimes the face ( almost always ) sometimes the body. and yet i havent figure out if i can blend faces as midjourney can. do you know about this? or anyone ? i have to develope this for a class, and i am starting to thing i choose a bad idea to generate i AI hahahahha
maybe, maybe not haha. there is an image section on the left inthe image generator, can you upload two images there? otherwise, Midjourney may be the way to go really.
Sort of. You do but only if the elements used in those images aren't protected by copyright. So ifyou create an image of Mickey Mouse, then no you can't use them. But create an image without trademarked or copyrighted characters / elements and I think you're good.
3136 pixels is more pixels tha 1024 pixels , so it is higher resolution by all measures. If you meant 3136 inches at 72dpi vs 1024inches at 300dpi then yes. But i didnt say that at all. More pixels is more pixels regardless of dpi
@@WadeMcMaster Yes, if you convert the 3136 at b72 to 300 DPI (which is minimum most POD and online stores is asking) you get image that can be printed on 8" square max without heavy pixelation.. 1024 at 300 is barely 4" so yes, your math is valid.
@@MarinaArtDesign does converting from 72ppi to 300ppi really change the image for the print? There are such services for conversion. I thought they were useless and just show high DPI without really changing the picture
If you'r working with models, Leonardo is not so great. Even after altering the negative prompt every time you get strange legs, too much arms etc. Too bad!.