This was SUCH a great video, no waffle, straight to the point, VERY good use of time I really appreciate the effort you put on this to give us the core information very fast. Thank you so much.
I haven't used Leonardo AI, but I like the look of the features it includes. MJ will no doubt start to incorporate more Standard Diffusion style functionality before too long. The competition between them should make for some exciting improvements, regardless of which platform you opt for. Thanks for your review.
Thanks mate, i agree, they are keeping each other sharp really. I'm still a bigger fan of midjourney but do appreciate Leonardo. Ultimately, Photoshop and MJ are the best combo imo :)
@@WadeMcMasterHere's a Merry question for you. I need to combine separate images into scenes by stitching together using gen fill. Which program do you think would give the best results in terms of matching styles and delivering faithful, high res combo images that look cohesive... Leonardo AI Canvas, Photoshop gen-fill or MJ vary region?
I prefer to use both Leonardo AI and Midjourney. I'm thinking of subscribing to the $12 plan for Leonardo AI and the $30 plan for Midjourney. I would like to mention some of the features that Leonardo has that Midjourney lacks, which are not discussed in detail in the video I watched. Things that Leonardo has that Midjourney doesn't: Image to image generator Controlnet Prompt generator Canvas feature Training your own model Things that Midjourney has that Leonardo doesn't: A describe feature A community feed gallery with a search function (the search function is temporarily disabled in Leonardo) More superior panning and zooming mode
Thank you for the input! I did mention the training of models and canvas, however - you can do image to image with midjourney but it's nowhere near as user friendly as Leonardo. I agree though that Leonardo does have more additional features which is cool
You are a crack 👍you are very kind for sharing this valuable information thank you look I have had some fascination to create a comic book about Greek mythology is not a very long project to say but the problem is that my drawing is not good according to your experience using midjourney do you think this tool can help me to make the comic book or what would you recommend me I would like to know your opinion god bless you 👏
It comes down to what you get best resutls with as some people may succeed with Leonardo, some with Midjourney. However, one advantage of Leonardo AI is that you can train your own model (thus fine tune a character more closely).
I would say the best thing I see is people going very detailed on the negative prompt (I might make a video on it) and also rendering the image at the same resolution as the model is trained on (the resolution is typically listed after you've chose your finetuned model - on the dropdown top right)
Thank you! And yes, that is true. Some platforms paywall the privacy and some don't, a point I didn't consider but very much a deal breaker for some people.
@@WadeMcMaster Thank you for your reply, its something I am contemplating on. I mean folks can just take your work any way due to a machine making it and not a human, it certainly makes me go hmmm
I've had some success with that, however the model you use does have a big influence also. So sometimes you may need to find the model that matches closest to the style, then apply that.
I have to say I'm loving Leonardo ai. I'm still learning the features but I really enjoy all the options and combinations. If I like an image but its not quite right and canvas hasn't quite got it, I'll polish it up in Photoshop. It is a platform which is being upgraded all the time. Having given all this praise, I do have a couple of nitpicks; animals are often hideous even with a ton of negative prompts. You get badly drawn faces, extra legs, ears, etc. I've also had annoying problems with human Images, extra fingers or disembodied hands etc. Weapons are another irritation... I'm sure they'll sort these glitches out, eventually 🙄
Which would you recommend that can use image prompts best? Example, I give it a portrait photo of my kid to put him in the park or running or whatever.
tough one, a portrait could work but may affect the image a bit too much. I think a better thing would be to train a model with photos of the child and use it in the prompt. I'd experiment with different images and prompts.
This was SUCH a great video, no waffle, straight to the point, VERY good use of time I really appreciate the effort you put on this to give us the core information very fast. Thank you so much.
Thank you! I do need to update this video though as a lot has changed. But it still holds mostly true. Have a great day!
Fantastic comparison video for everyone from beginners to pros bro! 🎉
Thanks mate! There's so many AI art platforms.nownita insane, good to compare and rank for people :)
Amazing stuff! Keep making!!!!!!!
Thank you! I'm glad you liked it :)
I haven't used Leonardo AI, but I like the look of the features it includes. MJ will no doubt start to incorporate more Standard Diffusion style functionality before too long. The competition between them should make for some exciting improvements, regardless of which platform you opt for. Thanks for your review.
Thanks mate, i agree, they are keeping each other sharp really. I'm still a bigger fan of midjourney but do appreciate Leonardo. Ultimately, Photoshop and MJ are the best combo imo :)
@@WadeMcMasterHere's a Merry question for you. I need to combine separate images into scenes by stitching together using gen fill. Which program do you think would give the best results in terms of matching styles and delivering faithful, high res combo images that look cohesive... Leonardo AI Canvas, Photoshop gen-fill or MJ vary region?
I prefer to use both Leonardo AI and Midjourney. I'm thinking of subscribing to the $12 plan for Leonardo AI and the $30 plan for Midjourney. I would like to mention some of the features that Leonardo has that Midjourney lacks, which are not discussed in detail in the video I watched.
Things that Leonardo has that Midjourney doesn't:
Image to image generator
Controlnet
Prompt generator
Canvas feature
Training your own model
Things that Midjourney has that Leonardo doesn't:
A describe feature
A community feed gallery with a search function (the search function is temporarily disabled in Leonardo)
More superior panning and zooming mode
Thank you for the input! I did mention the training of models and canvas, however - you can do image to image with midjourney but it's nowhere near as user friendly as Leonardo. I agree though that Leonardo does have more additional features which is cool
I am using Leonardo AI for my videos images and with Alchemy turned on these images are very good. Thanks for this review!
No problem! Alchemy is amazing :) this review became quite outdated quickly though, Leonardo has gone next level since then!
You are a crack 👍you are very kind for sharing this valuable information thank you look I have had some fascination to create a comic book about Greek mythology is not a very long project to say but the problem is that my drawing is not good according to your experience using midjourney do you think this tool can help me to make the comic book or what would you recommend me I would like to know your opinion god bless you 👏
It comes down to what you get best resutls with as some people may succeed with Leonardo, some with Midjourney. However, one advantage of Leonardo AI is that you can train your own model (thus fine tune a character more closely).
Do you have any tips on reducing errors on Leonardo AI? Preventing extra limbs etc!
I would say the best thing I see is people going very detailed on the negative prompt (I might make a video on it) and also rendering the image at the same resolution as the model is trained on (the resolution is typically listed after you've chose your finetuned model - on the dropdown top right)
With Midjourney, people can still see and take your art though, unless your paying the max for the service?
Brilliant video by the way. Very helpful.
Thank you! And yes, that is true. Some platforms paywall the privacy and some don't, a point I didn't consider but very much a deal breaker for some people.
@@WadeMcMaster Thank you for your reply, its something I am contemplating on. I mean folks can just take your work any way due to a machine making it and not a human, it certainly makes me go hmmm
That is definitely an issue with AI, I don't know if that will change someday or not
@@WadeMcMaster Exactly. I will continue to keep an eye on your content. :)
Does Leonardo work well with other styles? For example, “in the style of Gustav Klimt” works well in Midjourney
I've had some success with that, however the model you use does have a big influence also. So sometimes you may need to find the model that matches closest to the style, then apply that.
I use both and this review is spot on. I just which Midjourney had a built in upscaler like Leo
Thank you! I agree, if midjourney had an upscaler like LeoAi it'd be insane!
I have to say I'm loving Leonardo ai. I'm still learning the features but I really enjoy all the options and combinations. If I like an image but its not quite right and canvas hasn't quite got it, I'll polish it up in Photoshop. It is a platform which is being upgraded all the time. Having given all this praise, I do have a couple of nitpicks; animals are often hideous even with a ton of negative prompts. You get badly drawn faces, extra legs, ears, etc. I've also had annoying problems with human Images, extra fingers or disembodied hands etc. Weapons are another irritation...
I'm sure they'll sort these glitches out, eventually 🙄
Yeah, AI is so close but not quite at that truly free level yet. But I agree, Leonardo AI is moving fast!
I just like the one that understand simple promt, which one should i choose?
It sounds like Midjourney is better for you!
Which would you recommend that can use image prompts best? Example, I give it a portrait photo of my kid to put him in the park or running or whatever.
tough one, a portrait could work but may affect the image a bit too much. I think a better thing would be to train a model with photos of the child and use it in the prompt. I'd experiment with different images and prompts.
thanks for the quick reply!
@@WadeMcMaster
Do you get unlimited generations on either of them? Or are they both limited with credits?
Both are limited. Credits for Leonardo AI and gpu time for midjourney. However, Leonardo you can try for free.
@@WadeMcMaster that sucks, thanks though!
you forgot the fact that the time it takes to generate an image is way longer in midjourney
That may be true, but not my experience.personally. it could be I'm in Australia and am able to use it at lower traffic times
I look forward to MJ being able to generate accurate text.
Me too! V6 apparently is aiming at that
My mate owns Leonardo!
Awesome! He's done a great job with it
@@WadeMcMaster He's a very humble dude with big ideas.
is your upscale site paid site?
If you mean Topaz Gigapixel, it is a once off purchase
Hi
Hi!
Midjourney is way better
I prefer it too, I like both tho :)