🙋│image-help

[thin tinsel]

Right, so I'm going to say, go with EDGE TO IMAGE, on about 1.58.
Use the white on black image as the source
Do a couple few batches that can then be combined and layered up in canvas.

Prompt if you want it:
Ethereal crimson streaks cascade down the pristine white letters spelling out "ALL GONE." The main subject of this image is a hauntingly beautiful TYPOGRAPHY, a digital artwork. The vibrant dark-red paint appears almost sentient as it drips from the bottom half of the text, creating a striking contrast against the stark white text. Each drip is meticulously detailed, adding a sense of movement and life to an otherwise still image. This mesmerizing display of color and texture captivates the viewer, showcasing the artist's skill and attention to detail. Pools of paint gather beneath the text.

Trying to have the red 'paint' just covering the lower half of the text and then dripping off.

[thin tinsel]

How did you do it? What is your secret sauce?

[wise drift]

just phoenix. can knock out background later.

[thin tinsel]

I couldn't get the paint to fill the text, only drip off the bottom

[wise drift]

was a random one with auto prompt

[thin tinsel]

Ah OK.

[drifting swallow]

When adding text to an image via prompt. Are there any better ways over others to do it? Sometimes I use quotes ... sometimes I don't. Either way, most of the time it ads an extra letting for no reason, or changes a letter to a number.

[thin tinsel]

Quotes yes, do that.
Also front loading your prompt with the text you want.
Repeating what you want can help too.
If you're using Phoenix then you can be super detailed and say what you want where. But you cannot prevent from rogue text or logos.
You can however get rid of extra gubbins in canvas editor later on.

[drifting swallow]

thanks for the tip on using it twice... that might do it 😉

[thin tinsel]

If you have any examples to share, we can give you more tailored advice 🤗

[grand karma]

hi guys can any one help me and explain how i can create images of
angels from same image

[open echo]

Hey there! 😄
We’ll be happy to workshop it out with you

can you elaborate? You want to create angels in a similar style of an image? Or have these angels show up in another image?

(Also you don’t have to spam the same question in every channels 😅)

[storm jungle]

Logo's like these, you need to keep an eye on the blood running off the image, if you paste this over another image, the blood will look chopped

[chrome kiln]

@storm jungle lol I just put @ and @pure ivy automatically supplied your name

[pure ivy]

Ah, it seems my name has a certain allure, sì? It’s a pleasure to be so easily summoned! 😊 What can I assist you with today?
-# Our apologies if this message did not answer your question.

[chrome kiln]

senpai what would render "Ukiyo-e" artwork the best?
prompt: Intricately detailed, grainy Ukiyo-e woodblock print with bold, expressive brushstrokes, White koi fish and a black koi fish intertwined like Yin Yang and leaping out of turbulent water with waves curling around it. Red lunar in the center of the foreground. Dark blues, blacks, and white. A mountain, sakura tree and shrine in the foreground.

and here's my attempt

the ratio is wrong, completely

hmmm seems like a poster, not really traditional art @tall elbow

[tall elbow]

dame unos minutos y lo reviso

😅

[tall elbow]

How looks my first try?
Anime Flat Illustration - AlbedoBase XL
Low contrast
your whole prompt

[chrome kiln]

why it have those borders?

[tall elbow]

It must be because I chose a proportion that is not compatible with the trained models.

let me try again

[chrome kiln]

my only options are

• illustration
• stylized illustration

[tall elbow]

that it's not traditional

[chrome kiln]

albedo and illustration

but it's not the ancient wood carving style at all

https://tenor.com/view/cute-emoji-smiley-thinking-scratching-head-gif-17835192

[tall elbow]

did you use the element?

[chrome kiln]

yes

vision xl
medieval illustration
very close but it's .... just ... off

why are the waves off
it must be the model isn't trained in this

ok i'm just trying everything

[tall elbow]

Anime Flat Illustration - AlbedoBase XL
Low contrast
Woodcut 0.20 - Medieval 0.20

[chrome kiln]

Does the perspective change the actual render?

[chrome kiln]

well hells bells why didn't you tell me there was a woodcut

[tall elbow]

hahaha i did @chrome kiln... almost xD

Anime Flat Illustration - AlbedoBase XL
Low contrast
Woodcut 0.30 - Medieval 0.30

i change the order of the prompt... i think it could be better if you rewrite...

grainy Ukiyo-e woodblock print with bold, White koi fish and a black koi fish intertwined like Yin Yang and leaping out of turbulent water with waves curling around it. Red lunar in the center of the foreground. Dark blues, blacks, and white. A mountain, sakura tree and shrine in the foreground, expressive brushstrokes, Intricately detailed ancient chinese illustration,

Anime Flat Illustration - AlbedoBase XL
Low contrast
Woodcut 0.30 - Medieval 0.30

[chrome kiln]

HERE is what i'm trying to achieve (ignore the face, the person didn't redo the face)

[chrome kiln]

looks almost modern
#sigh

very close to it, no?

hmm this is extremely nice, but it's too ... polished? supposed to be flat colors, that roof back there has shading on it

[chrome kiln]

a so desu ne Zro sama

so you just gonna tease me or give me the prompt... um... both

[storm jungle]

tease for now, muahahahaha

[chrome kiln]

https://tenor.com/view/的9-gif-27299608

[storm jungle]

https://tenor.com/view/giggle-pfft-gif-26273029

Phoenix illustration <---------

Prompt:
Traditional Japanese Ukyio-e style illustration featuring a pair of koi fish, one pristine white and one glossy black, intertwined and facing opposite directions transitioning to forming a harmonious symbol of a black and white Yin Yang which sits behind the fish and leaping out of turbulent water with waves curling around it, structures, cherry blossom trees

You may need to generate a few, as mentioned, sometimes Yin does not want to Yang

[chrome kiln]

um

[manic root]

Hop over to #1108130218260041758 and take a peek at today's Daily Challenge! 🌟

[chrome kiln]

UM

oh you used phoenix

let me just...

i take 1 step forward and 2 steps back

[storm jungle]

Phoenix / illustration was all I used

[chrome kiln]

whispers i'm broke

[storm jungle]

Sounds like a tomorrow job : P

[chrome kiln]

https://tenor.com/view/thank-you-bow-michael-scott-the-office-gif-18616386

who invited the 3rd wheel?

[storm jungle]

Yin Yang and Yong

[chrome kiln]

lol

i mean ... it's pretty

https://tenor.com/view/movie-one-eternity-later-gif-7900643

so tomorrow you say?

[storm jungle]

This still phoenix ?

[chrome kiln]

I can't afford phoenix, of course. and now i'm out of points 🙂

https://tenor.com/view/auf-wiedersehen-tobey-maguire-maguire-tobey-auf-gif-5919412

[wise drift]

Considered adding “vector”

[gray dove]

This one is nice, can I ask what prompts were used?

[real lynx]

Good morning, the prompt is down there

A logo with the text "Frozz Void Studios" in a futuristic, bold font. Below the text is the image of a stylized ice with a scoop of Digital illustration featuring a stylized, geometric flame composed

Phoenix 3d render

[gray dove]

[uncut heath]

Is it possible to create this type of generation ( a character with different anime style ) ?

[wise drift]

Use same Character Reference, and different Style References

[uncut heath]

there no specific prompt structure?

[wise drift]

Prompt structure would be whatever you want them to be doing, background, etc. Character will handle the face, and Style will handle the actual drawing style. Can use Describe With Ai on an existing pic you wanna pseudo copy to get a good starting prompt.

[uncut heath]

ok i get it

[wise drift]

Fish eye lens prompt? Doesn’t seem to recognize it

[mild lintel]

how to generate images

here

i want to know that how to generate image here

[winged dew]

Does anybody know what shall I do to get a portrait in a similar style as this one

[open echo]

Hello there 😄

Leonardo.Ai is in fact a website with a full suite of visual tools. Simply go to http://app.leonardo.ai and you can start generating images for free 😄

And here's a quick 7 minutes tutorial to get you started: https://youtu.be/Rukln4nr_Z0?si=8ZctAfZZJqvpMVvl

[exotic oak]

is flux inside leonardo

[real lynx]

Hello welcome

Need help with an image?

[exotic oak]

Im wondering if flux is inside leonardo

[exotic oak]

hey

[real lynx]

I don't know what you mean by that.

[wise drift]

No, thats a different platform

[real lynx]

You want to know if something similar could be created?

https://app.leonardo.ai/

Our website is this one

@pure ivy borrar este link

[pure ivy]

I've sent a message to the moderators regarding your request to delete the link.
-# Our apologies if this message did not answer your question.

[exotic oak]

I mean if you can use the flux model on leonardo

[wise drift]

Flux is not a model. It is a platform.

[exotic oak]

it is

its open source

[wise drift]

Flux is the name of the platform.

[exotic oak]

no

[wise drift]

Yes.

[exotic oak]

type flux.1 dev on google

[wise drift]

It’s a platform.

[exotic oak]

its not a plattforn its a text to image model

[real lynx]

Check with the developers, they are the ones with the name in red.

[wise drift]

Does it have its own dedicated website that you linked?

[exotic oak]

yes but its black forest labs who created flux

[wise drift]

Then it is a platform.

[exotic oak]

if you say so

[wise drift]

That’s the definition.

[exotic oak]

Ai say its not a plattform its a model

plattform is something else

[wise drift]

I suggest you update your software.

[exotic oak]

this is flux 😬

[wise drift]

Again, I suggest you upgrade your software.

[exotic oak]

what software?

[wise drift]

Your software.

[exotic oak]

🤔💀🤷‍♂️❓

[wise drift]

Your files are corrupt and your software induces memory leaks.

[exotic oak]

Are you using a translator?

[wise drift]

Negative. I am the translator.

[chrome kiln]

QUESTION: I can't use "prompt magic" and "image2image" together in the old ui, the new ui doesn't have image2image, nor prompt magic

[manic root]

🌟 Hey everyone! Don't forget to take a peek at today's Daily Challenge over at #1108130218260041758! Dive in and see what's new! 🧠✨

[real lynx]

Hi

[chrome kiln]

oh you changed your pic

lol

[real lynx]

😅

I was just creating an image for that event

In this event, the logged model must be activated.

[real lynx]

Surely this is your mistake

That's because prompt Magic is not compatible with alchemy, it must be disabled.

Let's go to settings

[real lynx]

You turn off alchemy

Magic prompt appears

[chrome kiln]

https://tenor.com/view/youre-right-jabrils-i-see-what-you-mean-i-see-what-youre-saying-i-understand-gif-14299538170150748426

[real lynx]

It is image prompt the event

After that, you just have to play between giving more strength to the stick or the prompt.

To create an image

[chrome kiln]

https://tenor.com/view/hmm-interesting-intrigued-curious-thinking-gif-8065719971509630412

[real lynx]

If you have any questions, I'm still here ♥️

[chrome kiln]

ok you might be on to something LOL

[night vault]

while using prompt magic, you can only use image prompt one and nothing else works

[real lynx]

I had misunderstood

Tell me again what you want to do xD

[chrome kiln]

no i was the one who didn't understand at first 🙂

[real lynx]

It was a confusion of an ancient event

[chrome kiln]

we did it

[chrome kiln]

i got it

[real lynx]

Perfect and what do you want to do with that?

[chrome kiln]

well i thought it was for the official challenge, but it's over 😦

i guess

but i learned something

from a great teacher

https://tenor.com/view/hug-warm-hug-depressed-hug-gif-4585064738068342394

[real lynx]

XD

[chrome kiln]

why are yours looking like the stick figure still? you have your image too high in image2image

[real lynx]

I made these three images from another m

With premium tools

Keeps the stick figure with image content and I put a face on it with character inage

Then I just gave more strength to the prompt than to the images

Use the quling technique

That's why it has those paper-like ripples.

[chrome kiln]

https://tenor.com/view/surprised-sorprendido-shaquille-oneal-gif-23222312

[vocal portal]

The official challenge is the Daily. We do have other contests in various language channels. If you'd like to try one of them I can give you the link to them 🙂

[manic root]

Swing by and take on the fresh Daily Challenge over at #1108130218260041758! 🌟🎮

[broken vale]

Hey guys! I need a help. I am trying to realize this ai but the result is not good. Can you help me to writing the prompt? And witch one ai i can use?(Free)

[wise drift]

Use the Anime XL model. Click the little sparkle button next to Generate and hit Describe With Ai and upload one of the images. That will put your prompt as an AI description of the image. Then you can edit all the parts of the prompt that don’t describe the actual style and whatnot

[broken vale]

Thanks!

[night vault]

Describe tool help depicting character and color of the image. Re-create a style can be easy yo start hard to master topic. Depends on taste and what you are picky about. I can recommend @subtle gulch for realistic anime. He is also availabile in /mentoring if you want.

[broken vale]

Can i without premium?

[night vault]

Anyway, feel free to discuss what have you done with Leonaro and help me understand why result is not reaching satisfaction.

[night vault]

Describe is free tool too

[wise drift]

That’s the free way to do it. The premium way is better, though.

[still sparrow]

Anyone have any advice for replicating the exact look of a character into different scenarios for an image?

[wise drift]

Real or like cartoon?

[still sparrow]

Real

Honestly both, its something that I've been wondering for a while now 😅

[wise drift]

You’re wanting multiple poses of the same character in the same image? Or wanting character consistency between images?

[subtle gulch]

Does this look like what you want to achieve? (No premium needed)

[broken vale]

Prompt?

[subtle gulch]

Prompt:
Melainotype art of a cute anime female cybernetic wuxia, purple and white modernized wuxia costume, Key Visual, insanely detailed, bioluminescent, ultra dynamic pose, floral and butterfly embroidery patterns, light particles, floating fluorescent butterflies in the sky, extremely sharp focus, unique artistic composition, swinging a sharp katana, kirlian effect, whimsical magic, elaborate consistent background elements, Key Visual, cel shading, Rembrandt lighting, bokeh

The sparkling effect comes from the prompt: “Kirlian effect”

You can adjust the prompt and change the color and subject you want.

But the most important is the use of elements for better details.

[still sparrow]

Character consistency between images, so the character looks the same in different images

If that makes sense

[broken vale]

Ty my friend

[subtle gulch]

Here are the settings for elements.

[broken vale]

I Will try

[broken vale]

Ty again

[subtle gulch]

Just drop by if you need any more advice

[subtle gulch]

The easiest way is to use “Character reference” but that is a paid feature

[subtle gulch]

Certain models like AnimeXL tend to create similar faces with the same set of prompt so you may try that as well, but just bear in mind that variation is a constant in AI image generation.

[still sparrow]

Awesome, thank you!

[broken vale]

Is the same?

[thin tinsel]

Silioa has gone to bed now.
What is your question.

[broken vale]

Just realize same image of Zoro, same style

[thin tinsel]

What is the question? What do you need help with?

[broken vale]

If can be possible know the prompt of the style using for zoro

[thin tinsel]

Siloa gave you a prompt to try.
Did you try it?

[broken vale]

Yes, but i think is different

[broken vale]

Want see the prompt?

[thin tinsel]

Are you expecting it to be identical?

[broken vale]

Absolutly not, but if u want i can show you

[thin tinsel]

Ok show me

[thin tinsel]

Like this one.
#🙋│image-help message

[wise drift]

He didn’t give you prompt for the zoro one

[broken vale]

Yes but i try to recreate this one.

[thin tinsel]

I still don't understand what the issue is.

[wise drift]

Use Describe With Ai on the zoro one he posted

[broken vale]

Yes, thank you

[broken vale]

Wait, restart all.

[broken vale]

I want recreate this art style and i ask an help, can you help me to recreate this one?

[thin tinsel]

Purple one?

[wise drift]

He wants the general art style of both, but actually wants the right one (his name is zoro)

[broken vale]

Both, i think is the same art style

[thin tinsel]

Who is Zoro? Is this relevant information? I don't know who it is, so I was just ignoring that word.

[wise drift]

The character on right is named Zoro

[thin tinsel]

OK

[broken vale]

Is the name of the character, one piece anime

[wise drift]

He’s a character from the One Piece anime

[thin tinsel]

OK I don't know what that is.

[thin tinsel]

Do you like this image or not?

[wise drift]

I believe it would be Anime/Semi-Realistic, then all the lighting effects and the energy and whatnot

[broken vale]

Absolutly yes, but the main question was for the other two images

[thin tinsel]

Then why not keep the same preset , and just adapt the prompt Silia gave you?

[broken vale]

Because i cant do it

Sorry but i am the new own ia art

[wise drift]

@broken vale It is too difficult to translate?

[broken vale]

What?

[wise drift]

Are you having trouble with the English of it?

[broken vale]

Little bit , i am italian

[thin tinsel]

Siloa gave you the prompt and then even said, just change the colour and the subject.

[thin tinsel]

Woud you prefer to speak to mentor who shares your first language?

[broken vale]

Ok, if you dont likr talking to the translator

Like*

[pure geyser]

I was asking the same a few days ago
See
#🙋│image-help message

[thin tinsel]

You need to adapt Siloa's prompt.
change PURPLE to GREEN
change FEMALE to MALE

[pure geyser]

I think @muted canyon can help us , I think he is good in this

[wise drift]

How are those, Anthony?

[broken vale]

If i change the colors but the style of images that i shared (purple and green ) is different or same ?

[broken vale]

The art style is different

If you want i can shared you other image of this art style

[thin tinsel]

Are you saying that your green image is a different art style to your green image?

[manic root]

🚀 Ready for a new quest? Head over to #1108130218260041758 and tackle the Daily Challenge!

[broken vale]

No, i said that the art style of prompt that siloa give me, is an art style different from images that i share

[pure geyser]

Try semi realism

[broken vale]

Ok

[thin tinsel]

Ok , so when I asked if you were happy with the image you got ... I meant
Are you satisfied with the result you got from the advice and prompt you were given.

[wise drift]

Those were semi-realism that I posted. I think his original ones are 3D modeled

[broken vale]

It is a beatiful art style, but is not that i search

[thin tinsel]

Did you do this bit?
#🙋│image-help message

[broken vale]

Yes, i do same

[thin tinsel]

Can you see the difference between the original posted images, and the result the person got from Siloa's prompt?

[thin tinsel]

And are you able to describe what is different?

[broken vale]

I think si different only in art style, but i don't know this art style called

[thin tinsel]

And what adjectives would you use?
It is darker, it is more detailed, it is more like a drawing, it has more energy ... for example

[broken vale]

This is the result, is different to my images

[pure geyser]

See this type of realism is damn good .

[broken vale]

Sure

[thin tinsel]

That was made using AMIME XL ??

[broken vale]

The true diffents is the aura that have, and the anime realism art (i think)

[broken vale]

Yes i using that

[thin tinsel]

Is that was you think the differnce is? That the source images are more realistic?

[broken vale]

The main, yes

[wise drift]

His looks like Blender or some other 3D art program, that had camera filters on it

[thin tinsel]

ok thank you

[broken vale]

The ai gen of those images is midjourney

[pure geyser]

Whosoever expert in this and can help me to generate this in Leonardo plz let me know. Try if u can match this

[wise drift]

What was the prompt?

[broken vale]

Of the image on midjourney? Idk because i find this image on Instagram, of an ai artist

[thin tinsel]

I can't even see the difference, and now you say it's not even made with Leonardo, how can we confirm it is even possible to perfectly replicate

[broken vale]

I don't know if we can perfect replicate, but the style is different, Just that.

[broken vale]

This is in a style almost same, just need the aura for same result

[thin tinsel]

Ok, then best to ask for Siloa to return to the conversation.

He showed you #🙋│image-help message this one and asked if that was what you wanted

[thin tinsel]

What's an aura?

[broken vale]

This image it looks more like it like style

[broken vale]

The effect around the character

[thin tinsel]

Lightening?

[broken vale]

More or less, for example the effect in the first image is like water, in the second is like... Smoke i think

[thin tinsel]

ok

[thin tinsel]

In Silioa's prompt it says "kirlian effect". I think that's the effect

[broken vale]

Ah ok

[thin tinsel]

So you can make it stronger in the prompt, by using Capital Letters, or inside ()

[broken vale]

Oh thank you, i dont know It

[thin tinsel]

That's why I told you. 🥰
To help you improve your experience.

But anyway, let's wait for Silioa to return and he can offer some advice. 🙂

Thank you

[broken vale]

I think that this, Is the best result for this art style

[broken vale]

Ooooh so good

[broken vale]

Try to insert in prompt "roronoa zoro"

[thin tinsel]

This one I added Cosmic Retro element on 0.20.

[broken vale]

I dont know what is this

[thin tinsel]

Scroll up to see the images Anthony first asked about.

[thin tinsel]

Thank you. I am not as experienced as some others.

[broken vale]

Ah ok, i can add just 4 effects

[thin tinsel]

YEs, it's this which Siloa suggested here : #🙋│image-help message

We are the mentor team. YOu see the open notebook emoji next to my name?

The character will always change unless you take steps to keep characters consistent

[broken vale]

How i delete?

[broken vale]

For 4 coin

[thin tinsel]

You need to turn the toggle off

You can use the images anywhere

[broken vale]

But the prize is not more 4 coins

[thin tinsel]

Turn this off instead?

[broken vale]

Ok ty

[thin tinsel]

which two?

[wise drift]

@thin tinsel try Albedo, Anime General Style, Quality Mode, Photo Real, and your elements

[thin tinsel]

Thank you.

This is really not my style of image, that I usually make

[wise drift]

Waste like 2k coins to save someone their own subscription cost for Style Reference lol

[thin tinsel]

I don't know what you're asking me

[thin tinsel]

I could, but are they here to learn or to get an image?

[wise drift]

No im saying I did that lol

[thin tinsel]

No, I'm not. I told you I am part of the mentore team 📖

[thin tinsel]

Nice!!

[wise drift]

Those close @broken vale ?

[thin tinsel]

I'm off now. 👋
See you laterz! 🤗

[broken vale]

Amazing!

[real lynx]

Hi

[broken vale]

Oh mi god!

[wise drift]

Anthony, were the ones I just posted closer?

#🙋│image-help message Are these closer, Anthony?

[real lynx]

I like textures

[broken vale]

Oh mi god...

Mine... HAAHAHAHAHAH

[wise drift]

Dude that’s not even close to the same style as your original one you wanted lmao

Do you mean STYLE or do you just mean Zoro?

[broken vale]

Prompt?

[broken vale]

Style

[wise drift]

The style she is posting is not even remotely similar to the one you first posted and asked about

[broken vale]

Witch one?

[wise drift]

That is not even remotely close to the style she is posting now. She’s posting Zoro in regular anime style

[real lynx]

This is how it would be without premium tools

[wise drift]

Colorpop?

[broken vale]

Have you similar prompt?

[wise drift]

Prompt for what? The 8 or so images I recently posted is very close to your original style

[broken vale]

This

[wise drift]

Is that colorpop element?

[real lynx]

Negative prompt

anatomical distortions, anatomical inaccuracies, proportional inaccuracies, blurry, low quality, unclear features, incorrect features, mutation, deformed, disfigured, cloned, cropped, out of frame, out of picture, double head, multiple figures, signature

[real lynx]

It's just a prompt without any elements.

(Yayoi Kusama) .and Zdzislaw Beksinski styled miniature reenactment of a dramatic war scene. (polka-dotted dusk) as a backdrop. made with (mirrored glass and aluminum), A mix of (Vibrant Neon Pink, Electric Yellow, Acid Green, Cyber Blue, Fluorescent White) Tones, (imparts kinetic energy, against the inkwell black of the unknown). masterful color graded, Highly detailed surfaces, reflective and absorptive, contour and relief, hyper-detailed, .photorealistic, textured, luminous, subsurface scattering, specular highlights, ambient occlusion, dynamic range, volumetric light, microdetails, physically-based rendering (PBR), ray tracing, realistic depth of field, displacement and normal maps, dynamic lighting, Shot on 70mm, Depth of view, F/22.(((An anime-style medium shot of Roronoa Zoro from One Piece. He has his eyes closed and is holding his hands out to the sides. He is wearing his classic outfit, with a white shirt, a brown vest, and a black sash around his waist. The background is a serene landscape with mountains, trees, and a body of water. one- piece)))

Prompt

[broken vale]

@real lynx can you replicate that?

[wise drift]

That is the same art style

[broken vale]

Can you give me the prompt?

[wise drift]

Albedo model, Anime General style, the same elements that Willow told you, turn Photo Real on, have it set to Quality Mode (not Fast Mode)

(((kirlian effect))). an intense man (((covered in large fluorescent green flames))), background glowing green smoke, Key Visual, insanely detailed, bioluminescent, ultra dynamic pose, green lightning and green flame patterns, light particles, floating fluorescent green flames in the sky, swinging a sharp katana, whimsical magic, Rembrandt lighting, bokeh

[broken vale]

I try

[wise drift]

Cami far better than I am, so I’d listen to her on prompts.

Those are just the settings that I found were closest to your original image after burning several thousand of my tokens

[real lynx]

Similar or equal?

[broken vale]

Equal is better

[real lynx]

Question .. strange

Why do you want something similar if you already have that...

[broken vale]

It is not mine, just take from a IG Page

[broken vale]

Prompt?😍😍

[real lynx]

Lol

There is no prompt. It is someone's image from Instagram

[broken vale]

Nooo hahhhah just want the prompt, need to try replicate

[broken vale]

Anime there isnt

[wise drift]

You are on Illustrative Albido model?

[broken vale]

Albedo yes i think, i check it

[broken vale]

I cant chage

[wise drift]

Show me the top of the page

[wise drift]

Turn off legacy mode

[broken vale]

I cant do photo real, i havent Premium version

This is the result

Now better

Not good but better

Just... Try to learn the ai art

Are you good?

I am trying to replicate those

[wise drift]

(((kirlian effect))). an intense man (((covered in large fluorescent green flames))), background glowing green smoke, Key Visual, insanely detailed, bioluminescent, ultra dynamic pose, light particles, swinging a sharp katana, whimsical magic, Rembrandt lighting, bokeh

Try that prompt

[broken vale]

Because want replicate every anime character, i love it

[wise drift]

Try the prompt I just gave you

[broken vale]

Yes yes, was good

[wise drift]

Now this one

(((kirlian effect))). (Roronoa Zoro from One Piece) (((covered in fluorescent green flames))), Key Visual, insanely detailed, bioluminescent, ultra dynamic pose, green lightning and green flame patterns, light particles, floating fluorescent green flames in the sky, extremely sharp focus, unique artistic composition, swinging a sharp katana, whimsical magic, Rembrandt lighting, bokeh

[broken vale]

Not has first job

Nono, i am the manager of lan room

[wise drift]

A computer lab or cafe. LAN just means wired ethernet connection

[broken vale]

Yes

[wise drift]

How did the last prompt work

[broken vale]

(((kirlian effect))). (Roronoa Zoro from One Piece) (((covered in fluorescent green flames))), Key Visual, insanely detailed, bioluminescent, ultra dynamic pose, green lightning and green flame patterns, light particles, floating fluorescent green flames in the sky, extremely sharp focus, unique artistic composition, swinging a sharp katana, whimsical magic, Rembrandt lighting, bokeh

[wise drift]

Idk what he does there. Probably just sits there and makes sure people have paid or have memberships or whatever

[wise drift]

Yes

[broken vale]

Yes, more or less, is my work

[wise drift]

More like “attendant”

[broken vale]

I am the entrepreneur

Yes, very small

Italy

[vocal portal]

why are you asking people where they live

[vocal portal]

and you. why is this member's line of work your business?

[manic root]

🌟 Ready for a new test? Swing by our Daily Challenge in #1108130218260041758 and give it a go! 🧠

[wise drift]

Was their conversation. I was assisting translating.

[molten pagoda]

hey all, is there a way to see old prompts for Canvas Editor?

[molten pagoda]

is there a way to force white background for canvas editor? it keeps giving me random dark backgrounds that makes bg removal a big task.

[night vault]

you did prompt for white background right? did you check if you canvas area has a bit of black that model thinks it need to fill in?

[night vault]

also background removal could be done from image that generated by Leo. check option near image upscale or check transparent background from get go.

[chrome kiln]

@pure ivy give me a good idea for greyscale epic pic

[pure ivy]

Ah, for a greyscale epic pic, imagine a grand scene of a knight in shining armor standing atop a cliff, with a majestic castle in the background, shrouded in mist. The interplay of light and shadow will bring out dramatic contrasts and depth. Buona fortuna with your creation! 🎨✨
-# Our apologies if this message did not answer your question.

[chrome kiln]

oh that's really good

[pure ivy]

Grazie mille! I am pleased you think so! Buona fortuna with your creation, my friend! 🎨✨
-# Our apologies if this message did not answer your question.

[pure garden]

Hello everyone! Im here to ask you for a small help. Is Leonardo available for creating sprite-based characters like this for example? https://cdn.leonardo.ai/users/7cf5bcb5-3fe4-4322-961e-5e053687bc40/initImages/6b40824a-a68e-4c05-957c-a3aa05f993d7.png?w=512 and if so, could you please help me how to get it work? I have premium ofc

[night vault]

do you want limbs to be separate for use in game or animation asset ?

[pure garden]

exactly!

[wise drift]

Looks like vector art

[pure garden]

doesnt mind if its vector or raster, its created by me by my hand, but I need to generate separated character parts, so I posted just a clarification and imagination of wanted output

[pure garden]

yeees absolutelly! How I can get this output please, any hints pleae?

[subtle gulch]

To be honest, it’s quite hard to get the AI to create a perfect sprite sheet ready for rigging. You’ll have to be prepared to do a lot of trial and error and post-editing to make it work.

I basically just literally ask for it with the Phoenix model:
A 2d body parts sprite-sheet side-view animation character design sheet of a hero with head, arms, hands, upper body, lower body, legs and feet all separated

[pure geyser]

@subtle gulch
Check this👇
#🙋│image-help message

help me to generate this in Leonardo plz let me know. Try if u can match this it's more like semi-realism character or human like body effect

[broken vale]

Hi Siloa! Are you on?

[broken vale]

Parli italiano?

[pure geyser]

@subtle gulch This image is insane teach me if you can

[wise drift]

If you’re free mode, you don’t have Photo Real option

[subtle gulch]

Yes I am

[broken vale]

Amazing!!

[subtle gulch]

I believe I’ve shown you one already before.

[broken vale]

I can't try now because ended the coin, but if u want i show you the result and if you want, help me

[broken vale]

Prompt?

[subtle gulch]

Those are what he found on the Internet, just like those you showed us before.

[broken vale]

This is the objective

[broken vale]

This is result for now

[wise drift]

Dude, you need Photo Real for it

[broken vale]

Mh... With pro?

[subtle gulch]

It’s getting there but don’t expect the outcome to be comparable with free version.

[wise drift]

The ones I generated yesterday that matched the style you want were done with Photo Real and the same prompts/settings I gave you

If you get the basic membership, you can also just use Style Reference

[broken vale]

@subtle gulch this is not with Leonardo

[subtle gulch]

You can get that realistic touch with universal upscaler as well

[manic root]

Hey everyone! Don't miss today's Daily Challenge waiting for you in #1108130218260041758! 🌟

[subtle gulch]

I know very well that’s not from Leo

[broken vale]

Can you show the image?

[wise drift]

For reference, I spent HALF A BASIC MEMBERSHIP WORTH OF TOKENS yesterday trying to help you replicate something you could do with Style Reference as a member

[wise drift]

I forget about it because it’s not on app lol

[subtle gulch]

You may not get anything that close to the original with Leo as it’s trying its best to comply with the copyright laws in Australia.

[broken vale]

Ah... Sorry...

[broken vale]

Mh.. ok

[wise drift]

Those are the same style

[subtle gulch]

If he wants THE Zoro from One Piece, then these are not what he wants.

[broken vale]

I dont knew that premium have been tokens, Sorry mate

[wise drift]

I also posted ones with zoro in that style. It does take zoro’s full name in the prompt

[subtle gulch]

Leo’s model does have Zoro’s data but it’s not the latest one-eyed version.

[wise drift]

If those were made in Midjourney, why not just go ask the midjourney server

[broken vale]

Bcause the mid is not free, i cant use not free

[subtle gulch]

I’d say it takes post editing even after generating from MJ.
MJ doesn’t just produce images with that level of details right away from prompts.

[wise drift]

It’s not one of the anime I like. Interestingly enough, after Willow and I managed to get the style down for him, Cami came in and posted just regularly old anime preset images with actual Zoro in them and he was like “yes! That’s it! That’s what I’m talking about!” And I was like “come on wtf dude that’s not even remotely the same, it’s just Zoro “

[broken vale]

If you want in private i can send the link of artist ig

[subtle gulch]

And I can tell you some MJ users in fact uses Leo’s universal upscaler as well.

[wise drift]

Why don’t you just go ask the artist then

[broken vale]

I asked, nothing...

[wise drift]

Go ask the artist how he made the pics you like. Why have others guess if he already knows?

[broken vale]

Want the ig Page?

[subtle gulch]

Just know that that level of images don’t come free

[wise drift]

Sure.

[broken vale]

If u want u can watch PM

[broken vale]

Watch PM

[wise drift]

Dude. He says it was Midjourney Neji. AND he has a link for his own discord server

[broken vale]

Link dont work

[wise drift]

Then you’ll have to wait until he updates it

[subtle gulch]

MJ/NIJI’s database has been reinforced with loads of copyrighted content so if you’re really aiming for exact replicate, that is probably your best bet.

Just know that you can still get into trouble if you try to use them for commercial purposes.

[night vault]

yeah I’m struggling upscale anime line art. maybe this is the correct way to use it.

[broken vale]

Is not commercial, just for fun

[broken vale]

Amazing! Is free?

[broken vale]

Omg!

[subtle gulch]

Unfortunately no, you’ll need the paid version to reach that level of details.

[wise drift]

What was it you did?

[broken vale]

Unlucky! Hahahah

[subtle gulch]

These are the free version ones.
May need some editing to fix.

[subtle gulch]

It’s just AnimXL with Raytrace and a cocktail of elements

[broken vale]

I Understand

[broken vale]

Incomparable

[wise drift]

Raytrace? What elements?

[subtle gulch]

I’m using my Mobile app with the default legacy mode. There is an option for RayTrace using quality mode (it’s called alchemy in legacy mode)

And here is the cocktail of elements

[pure geyser]

?

https://tenor.com/view/cat-no-nonono-noooo-cat-no-gif-13597132752790194338

[subtle gulch]

#🙋│image-help message

[pure geyser]

Share me the prompt to know better

[wise drift]

Never tried negative values on elements before. What’s the raytrace do?

[pure geyser]

I can't use image reference for someone image everytime. I need a blueprint of this prompt to create more of the characters of this realism/photoreal type

[subtle gulch]

Raytrace is just one of the presets you can find while using other models like Kino/ Lightning. It focuses more on creating light and shadow.

[wise drift]

The negative values basically forced it to look more realistic?

[subtle gulch]

The photorealistic effect is created after using the universal upscaler. It’s not the prompt itself.

[subtle gulch]

It’s not exactly that. It is quite hard to explain the negative elements… it’s probably better to explain it in a stream lol

Anyway, without lengthy explanations, it forces the AI to generate more details in the expense of clarity and prompt coherence.

But by doing so, it generates a more realistic/ detailed image. The extra noise created can be edited out or improved through upscaling.

[real lynx]

I speak Spanish

Hola encantada de conocerlo ,buenas tardes

[broken vale]

Claro!

[broken vale]

Have you ig Page for your ai?

[broken vale]

Have you ig Page for your ai?

[subtle gulch]

I have an 𝕏 account, not IG

[broken vale]

I havent it

[subtle gulch]

If you want realism, adding “realistic/ photorealistic” at the beginning will make a difference.

[pure geyser]

Do you use image reference to create this? I ask to create this with prompt(as you know how to prompt)
I am not getting with describe with ai.

[subtle gulch]

No image reference, prompt only

[subtle gulch]

Describe with AI won’t usually give you the anime character you need

[pure geyser]

I know

[subtle gulch]

If you have a paid account, it would be easier to do by just giving the anime character’s name a heavier weighting.

[pure geyser]

That's why I was asking for the prompt you use

[subtle gulch]

I actually already shared it with Anthony just yesterday here I believe.

[subtle gulch]

You want the naruto one?

[pure geyser]

Yes

[subtle gulch]

Photorealistic Overexposed full body close cropped shot of an edgy cool beauty cosplaying as (naruto posing), whole body engulfed in red fiery flame like energy, head tilted up, Kirlian effect, overexposure portraiture, dynamic pose, intricate details, well defined radiant skin texture, volumetric cinematic lighting, Rembrandt lighting

And these were the elements to go with it

[pure geyser]

https://tenor.com/view/excited-happy-gif-15782106255560805477

[subtle gulch]

Happy prompting.
I need to go get some sleep now.

[real lynx]

Super abandoned sometimes I remember and upload things

I prefer to sleep and watch anime

And pending task to do exercises

Every day my treadmill looks at me and asks me when are you going to stop using me as a clothesline?

[shut token]

Tell me some prompt for instagram blogging for games

[dim zodiac]

Hey, I've been seeing a trend of people coming in here and DEMANDING prompts or certain tweaks on images, or aid in describing a certain image.

Can we all remember that the people helping in here are volunteers, and trying to help users get the most out of Leonardo

A little bit of respect, let's get some "please and thank-you's" for the helpers

[manic root]

Hey everyone! Don't miss today's Daily Challenge waiting for you in #1108130218260041758. Give it a try! 🧠✨

[snow pasture]

And sorry if this image is kinda horrible

[open echo]

@snow pasture hey there!

I’ll just go ahead and remove the image.. because, yeah it’s kind of horrible 😅

What kind of prompt have you tried so far for the security footage effect you are looking for?

Anyhow I can say Leonardo Kino makes great “camera” footage

[snow pasture]

Oh sorry again about it , next time i try send better examples 😅

[snow pasture]

I think i should've use the camera as an effect

I will try that

[chrome kiln]

This isn't pointillism help me please

@real lynx

[real lynx]

Hi

Want a similar pointillism image

[chrome kiln]

yesssssssssssssssss

[real lynx]

Sry my spanish 😅

I haven't done anything for the event yet

I'll see what I can create ✌️

And I teach you

[open echo]

Use the Leonardo Pheonix model 😄

Yup! Leonardo Phoenix 😁

Yes! 😁

On your web browser, visit the website “app.leonardo.ai”

[chrome kiln]

I still feel this isn't pointillism

[night vault]

I’d like it a bit more yeah. I remember last poinilism I need to explore other models and style combinations.

[night vault]

yeah better result 🙂

[chrome kiln]

kind of directly the pic

[real lynx]

Hi

[real lynx]

It took me quite a while

[chrome kiln]

what did you create?

[real lynx]

Mix a normal image

[chrome kiln]

lol of course you made something adorable

[real lynx]

I uploaded this to image style

[chrome kiln]

this was your image2image?

😮

[chrome kiln]

wait, you mixed the two?

[real lynx]

Yes

Prompt.+imagen style

[chrome kiln]

you're an excellent teacher. i learn so much

[real lynx]

I'm so glad you're learning.

[night vault]

lovely, you are too good at this

[storm jungle]

@chrome kiln

If you want a prompt only, you can play around with this.... The following prompt created the image exmaple you see here:

A bright vivid intricate colored pointillistic dot painting of a colorful butterflies in the garden all entirely created in perfect painted dots

If you find at times a building or the sky looks like it is painting instead of dotted, you can just note it at the end of the prompt like this:

Prompt:
A bright vivid intricate colored pointillistic dot painting of a colorful butterflies in the garden all entirely created in perfect painted dots, all butterflies and the sky is also created in perfect painted dots

Phoenix / Dynamic

[pure garden]

Hi! Ive somehow created this image and I would like to say something like "make sprite sheet of this character - head, legs, arms separated" is it possible with leonardo?

OR

Is possible to convert this to 3D object somehow?

[lucid oriole]

So i found this image on pinterest, and i'm kinda interested to make the similar image in Leonardo AI. But i don't want to use the image guidance, can i actually make the same style on Leonardo?

[real lynx]

@small sierra

Welcome

You can share the prompt here or if you want a personal channel you can use the command /mentoring

[manic root]

Jump into today's Daily Challenge and get those gears turning! Head over to #1108130218260041758! 🧩🚀

[storm jungle]

Hey there, I think I have just the prompt for you, see below the prompt, just change the wording to suit what you need, such as hair color, clothing and background etc.

(image examples are from the prompt below)

3D Toy Girl: A kawaii chibi girl with aqua hair at the center wearing a panda onesie while drinking bubble milk tea, the design embodies a Bauhaus-inspired aesthetic with a vibrant color palette. This high-definition, ultra-detailed digital art piece utilizes a boundary-less design, created with tools like C4D, Octane rendering, and Blender. The artwork, in its full-body close-up view, is set against a minimalist background, accentuated by global illumination and brilliant lighting effects.

Done using Leonardo Vision / Dynamic

[wanton belfry]

well sort of, but it might look something like this

[wanton belfry]

it deals with creating permutations and combinations so I dont think so. Even if you by miracle get something similar it still will have alot of differences

[lucid oriole]

I have already made this Image using this kind of prompt:

A Colorful illustration of a anime character wearing a black and white sci-fi jacket with fully accessories, eye-glasses, a cap, large boots, posing on a white background in chibi art style

But even i'm adjust the prompt to get unique pose or unique clothes, like the pinterest one. It still difficult to achieve that pinterest kind of style ._.

[night vault]

use style reference. it can help with clothing design and color

[chrome kiln]

ok @night vault help me with this one, and I can't use Phoenix: PROMPT - product advertisement: can of "Low Sugar Water"

@storm jungle

it's a joke, my joke, i thought this would be funny

I'm getting closer?

[storm jungle]

Why you can't use phoenix?

[chrome kiln]

i only get 150 tokens per day 😦

[lucid oriole]

Orrait, i'm gonna try it later

[storm jungle]

Well to get the text right with other models, you going to need a lot more than 150 tokens, phoenix does text perfectly, with the right prompt, you will get it first go.

[chrome kiln]

whispers, this is from a different thing, don't tell anyone

[storm jungle]

HEY EVERYONE!!! 😛

[chrome kiln]

https://tenor.com/view/shhh-shush-silence-nose-gif-17895433

[hardy sentinel]

hello 🙂

[woeful falcon]

Hey folks! I need a little help with prompts: Photoreal always drawings a pants joined with a boots together like as the attachment. How i should build my prompt for fix it?

[dim zodiac]

Can you share your prompt so that we can suggest edits?

[woeful falcon]

The early foggy morning, dim pre-dawn light. Full body, high-angle security camera feed. The 20-years old French girl in the vintage black slim boot cut leather pants and the brown knitted longsleeve. Her leather pants should be in the classic five-pockets style. Her leather pants should looks very thick and very old. She is sitting on the wooden bench on the top of a tall hill. Long flowed dark brown hair, big striking blue eyes. Her face features a long shape, with small-sized cheeks. The extremely big blue eyes. The extremely delicate facial features. The long soft dark brown hair. The rich hair color. The very slim skinny body. The easy happy facial expression. The autumn, October. The snow-capped mountains over the fog line far on the background. The 1980s action movie mood. UHD, 64K, highly detailed, CCTV still, full body, full body view, RAW photo.

[dim zodiac]

My suggestion knowing nothing of the prompt would be to describe the boot style. "Ankle boot", "riding boot", "galoshes", "thigh-high boot", etc

Yes, so in this prompt the boot is only barely mentioned

Awesome prompt by the way!

It is perhaps a little wordy for Photoreal, you could take out every "the" and likely get better prompt adherence

[woeful falcon]

I had try one more prompts with same result and i used image guidance with this pic

[dim zodiac]

That one the boot and pant are separate at least!

But yes, remove every "the" and spend at least a few words describing the boot itself. Right now it's only mentioned in relation to the pants

[woeful falcon]

Thanks! I'll try it

[dim zodiac]

It looks like your input image might be causing issue for you as well, because the model in the image is wearing pants that are very similar colour and material to the boots

Your prompt will work quite well in Phoenix, and then you could use that to generate a cleaner image in Photoreal after, if you want more control

[woeful falcon]

It works! Thank you!

[dim zodiac]

Perfect, glad to hear

[karmic parcel]

i am not able to generate images -- error Content moderation filter

#🙋│image-help can somebody help

[manic root]

Hop over to #1108130218260041758 for today's exciting Daily Challenge! 🎯✨

[warped locust]

Does anyone have a prompt for generating medieval paintings? old canvas paintings, I want something more realistic

and a model that would fit well with this type of art.

-# Oil painting of "The Three Musketeers, Athos, Porthos e Aramis" Medieval painting, canvas textures, in canva, Realistic paiting, ((Medium shot/distance))

[wanton belfry]

realistic and and oil painting might clash and then it becomes one or the other

rather try adding style of oil painting over it

yea much better, I added a oil painting style on it and there you go

this is generated in Kino Cinematic, style ref, (Oil painting:1.8) of "The Three Musketeers, Athos, Porthos e Aramis" Medieval painting, canvas textures, in canva, Realistic paiting, (Medium shot/distance). More or less same prompt just removed and added some brackets

[warped locust]

@wanton belfry

[warped locust]

yeaaa

In that same style as I wanted

[wanton belfry]

might pick out a image reference of the sort oil painting you like for it and use it in Kino

no need for alchemy unless you want too much details, it might make it look like their face looks all the same and weaken style

[warped locust]

yeaa, Thanks!!

[wanton belfry]

no problemo, have fun

[night vault]

whatsup, are you getting desire result yet?

[dim zodiac]

What are you trying to get? If you can share your prompt and whether you're using an image input?

[wanton belfry]

possibly using a blacklisted word or conjunction, might need to post the prompt to check what exactly

[red abyss]

#help............ is there any option to download multiple images at once or select multiple images to download at once?????,,,,,, because Leonardo's download system is too painful, i have to select 1 by 1 or drag over every image to select multiple images

[karmic parcel]

Thanks! It helped

[wanton belfry]

you have to click on top left on the images and it'll look like this

and then click download

this is when you try to download from feed

click on select and select images you want from feed

[chrome kiln]

no, i assume phoenix is the only thing that can generate actual readable text

[night vault]

yea, that’s why it was advertise as feature

[plucky shale]

hello, I am an architect and I would like to have the possiblity to create hand sketch/artistic images from photorealistic, rendered images. I need to retain a high level of detail and similarity to the photorealsitic image. I am wondering what would be the best workflow in leonardo to achieve this? I have attached some images to show an example of a photorealsitic image and an example of an artistic image to demonstrate the outcome i would like.

[real lynx]

Does something like this help you?

[plucky shale]

its on the right track but too much detail lost, the trouble i am having is trading off too much detail in order to give the ai enough room to change the style

[real lynx]

I think what you are looking for is canvas editor

It keeps changing

[wise canyon]

Hi Shamus, you might also find "Edge to Image" and/or "Depth to Image" (in image reference options) useful tools for getting the results you're after,
you can play with the strength slider, prompt, and presets to dial in the way you want your image.

this is a quick "Edge to Image" example:

[plucky shale]

wow very nice !

[wise canyon]

Here are the setting I used as an example:

[plucky shale]

this is great ! seems to retain lots of detail, I will give it a try - thanks

[wise canyon]

Yes, "Edge to Image" helps keep the lines the same quite well,
"Depth to Image" could work well too, as it creates a depth map of your input image 👍

[gray crypt]

hi all, i would need a little help with an image, i want an image of a whale in the clouds, not the sea, but it keeps generating me the whale over the sea. here is the prompt:
A surreal image of a whale swimming through a stormy sky as if it were its natural habitat. The sky is dark and threatening, filled with dense, swirling clouds in shades of deep gray and purple. The whale glides smoothly among the clouds, its body blending seamlessly with the turbulent cloud formations. In the distance, a forming tornado is visible, adding drama to the scene. The whale appears in perfect harmony with the celestial environment, as if the air were its ocean. Raindrops and mist surround the cetacean, softening its outlines and further integrating its presence into the stormy sky. Lightning occasionally illuminates the whale's massive form, revealing its true colors against the dramatic backdrop.
I put negative prompt: (((no water))) ((no sea)), the whale it is not in the sea, not over the sea,

[real lynx]

In Phoenix

Big storm, A whale flying through the sky amidst fluffy white clouds. The whale is large and is flying with its mouth open. The background contains more clouds and the horizon. The overall image has a

[remote beacon]

Why are we no longer able to delete images unless we go to Premium?

[compact bone]

sorry but how the hell does your application work? I simply write: greek god apollo portrayed by chris hemsworth talk with orlando bloom, and this is the image...

[simple robin]

What is the best preset to generate images where the hands are relistic ie.. 5 fingers holding something instead of the hand looking all weird etc? Portrait perfect?

[misty tiger]

I will also try it now

[hardy sentinel]

the best strategy is to hide the hands

[misty tiger]

No, he doesn't look like Chris Hemsworth: @compact bone

[hardy sentinel]

like this, you don't see any hands

[simple robin]

yes i delete the images that comes out bad, not all of them are

[manic root]

Hey everyone! 🌟 Don't forget to take a peek at today's Daily Challenge over in #1108130218260041758! It's a fun way to test your skills. 😊✨

[hardy sentinel]

well, you don't need to delete them, you can edit them later too 🙂

[simple robin]

you mean crop them?

[hardy sentinel]

yes, or canvas editor

[simple robin]

How would you use canvas editor?

[hardy sentinel]

cropping or using GIMP manually sometimes is cheaper if you have time , canvas editor if you have pc

[hardy sentinel]

are you using pc or mobile ?

[simple robin]

pc

[hardy sentinel]

ok, wait

this is the home website of leo

why is my screenshot not uploaded

please select number 6

[simple robin]

Just wondering when they are going to upgrade the app to fix hands

[hardy sentinel]

well, I am a user like you

[simple robin]

Interesting, is that your desktop??

[hardy sentinel]

yes, months ago

[simple robin]

ah, wait, now I see where that is. thanks

[hardy sentinel]

yay , you are welcome

[simple robin]

May I ask what you use your images for? I am looking for ideas

[hardy sentinel]

mostly just for fun, I make mod for games too

[simple robin]

Cool

[hardy sentinel]

and how about you?

[simple robin]

I am good. thank you. Just looking for ideas starting out.

[hardy sentinel]

I was asking what you are using your images for ? Since you have asked me.

[simple robin]

yea, thats what I dont know. doing a bit of research

[hardy sentinel]

why ?

[simple robin]

Im a stay at home mom with internet access.

[pure geyser]

does leonardo not know the characters of demon slayer.
What a crap is this

Giving some random female character, I ask for Zenitsu Agatsuma(a demon slayer character)

[karmic pulsar]

Lol

[hardy sentinel]

obviously not

[pure geyser]

Then what is this
https://cdn.discordapp.com/attachments/1170046881183834192/1283824105162018907/Screenshot_2024-09-12-21-46-43-55_40deb401b9ffe8e1df2f1cc5ba480b12.jpg?ex=66e465de&is=66e3145e&hm=b01b761a2aa5a8389300214e08019c74c565cb5eb1fd2bffae2c6d35f4a571d0&

[hardy sentinel]

I don't know

I have said it is obviously not what you wanted

[hardy sentinel]

I am a user like you. If you want to express your anger please contact leonardo employers and employees

[dim zodiac]

Actually don't do that. There's a few simple explanations why you can't call out a character by name

Firstly, Demon Slayer is relatively new compared to training data that would have gone into the models

Secondly, when it comes to training with intellectual property, Leo has always gone with models that exclude specific IPs

So you can get yourself an image in the style of certain anime, but not necessarily characters by name

[real lynx]

You can create a similar character even if you don't know Leo.

I used the description of a zenitsu internet doll

[dim zodiac]

Absolutely! There are tons of ways to achieve it, I was just responding to Striky's anger that his 5 word prompt that called out an anime character didn't give what he was after

[simple robin]

Is it possible to generate logo images without background, like in png files ?

[real lynx]

You have an example

A picture is worth a thousand words

I assure you that yes, but I want to see about creating something similar for you.

[simple robin]

I want to take a jpg like this, vector and remove white background to be with no background

[real lynx]

If something minimalist

Do you want the logo to say something?

Or just a cat?

[simple robin]

Just a cat

what you call an image with no background?

[real lynx]

For example

For example

[simple robin]

I have to remove the background alltogether if i want it on a t-shirt

[real lynx]

That may help if you already created an image

[simple robin]

yes, like the last one

[real lynx]

A minimalist logo of a black cat with the text "Buenos días" in a modern sans-serif font. The cat is in a relaxed position, with its eyes closed and its body curled up. The text is placed under the cat. The logo is designed with a monochromatic color palette, using shades of black and white. The background is transparent.

Phoenix is our tool for logos and posters

Say good morning and good afternoon in Spanish

Anything that is a letter and you want to appear in your image has to be in quotation marks.

[simple robin]

wow. I have so much to learn

thanks

what preset is best for vector?

[compact bone]

It seems incredible to me that this AI does not understand how to insert two distinct people into an image who are talking and who are aesthetically as I say, if I want two famous actors who are talking with ancient Greek clothes, a very simple scene, it does not understand anything, it puts only one, or three... but what the hell is it doing, I'm wasting all the tokens I bought....

Is there any support? Someone who intervenes for a paying customer?

[simple robin]

I had that, not doing transparent background.

[real lynx]

Yes, because I made a 3D render

[simple robin]

Ok, know how to remove now

[real lynx]

Vector desing

[simple robin]

thank you. will try that

[real lynx]

There are also other presets that if you add elements to them

[simple robin]

this more what I am aiming for

[real lynx]

Perfect ☺️

I'm going to eat any questions you have, let me know and I'll come back later to answer 👋

[manic root]

🌟 Hey everyone! Don't miss out on the Daily Challenge waiting for you in #1108130218260041758! Swing by and see if you can tackle it! 🧠✨

[proper crystal]

Why isn't it removing the text but adding text onto it? It worked before and now it isn't

[topaz hawk]

Hi people, I am so used to the legacy version of Leonardo, I get lost a bit.
Where do I find the functionality to create from a reference image?
As in: I create a concept and leonardo fills it with some life.

[hardy sentinel]

Hi, you can click at the top right to go back to legacy mode

[thin tinsel]

If you like to stay in Legacy, then you can use image to image in there.
But will be different depending on whether you're a paid user or not

[topaz hawk]

Oh no my new digital friends, I enjoy the current version a lot. Got some good settings setup for expected results and it does the job right. Except now I have a separate task to work from an image delivered by user. While using "my" settings with the current Leonardo.

[hardy sentinel]

wait I have a screenshot

[hardy sentinel]

this is the new interface, please click number 2

[thin tinsel]

I think everything I said remains to be true whether you prefer legacy or new UI

[hardy sentinel]

thereafter a pop up will open where you can select which mode of image reference you want to use

[thin tinsel]

What do you have as reference, is it a sketch or photo, or other computer generated image

[topaz hawk]

k, payed user feature I assume 🤔

[topaz hawk]

nothing yet but I have to, to be specific, design an environment for a game with specific areas - then Leonardo can make things look neatly from my concept

[thin tinsel]

Ok, I don't think we can advise which option to choose without knowing what the source is.
As for free or paid. I don't know which ones are available.
Maybe StevenP knows?

[hardy sentinel]

please show me your screenshot

[marble cape]

ok so im having a little bit hard time on something as you can see this is my orginial artwork i did on my ipad on left and i generate it which looks like this. there couple of things first dont see the full branch as you see and second just feel that the herbs by looking at it not like planted on the ground

I have others to but i think the one on the right seems to fit better i guess

this is my original one i did on my pad

[topaz hawk]

there is none, but it would look like this, a tad bit better.
Understand, i am seeking the "make image from exampel image" functionality, before i actually the work, gotta know if the software is capable of doing this (and i remember it is). So I would draw out a nice map with all the stuff needed and porperly colored and adding textures etc so the ai easily detect which part is what and then do the ai magic and actually make a nice image from it 🙂 So this mockup is really a quick "paint thor together" base - not the actual image that would be used. But I assume the function i am seeking is the one oblivian is using in his example: Make an image from an image

[hardy sentinel]

ok, i will get back to you maybe tomorrow. Sorry I am busy right now

[thin tinsel]

Depends what output style you want, how much detail and how much variation.
This is with image to image.

[proper crystal]

I assume no one knows how to remove a little bit of text?

[thin tinsel]

Yes, but it's probably going to be another 40 minutes spent talking about it and I need to sleep.
So if you're OK to wait then someone will pick up your question later.
Small tip .... don't put the word TEXT in the prompt.

[topaz hawk]

yeah, image to image! That's what I was looking for - how did you get there with the current leonardo?

[thin tinsel]

Steven showed you in the screenshot.
#🙋│image-help message

Be prepared to do some experimenting with different models and styles, and weightings, and a good prompt will serve you well.

[topaz hawk]

Don't worry, i am around leonardo for a while, just used legecay for too long, gotta find where things moved to. Thank you, this is what I was looking for. Request solved

[dim zodiac]

Heya Vallywally! Best way is likely to take it to the Canvas Editor

I would use the sketch to image mode and paint over the text you don't want with a colour eydroppered from the background

[proper crystal]

It is clearly in the canvas editor

[marble cape]

yeah im stuck

[real lynx]

It reminded me of the old pokemon games

[marble cape]

...

[real lynx]

😅

[marble cape]

...red

so what how do i fix the herb to be more like on the ground and how do I fully see the branch thing that you guys see

?

[real lynx]

It doesn't really look like grass, it looks like crowns, I would look for new seeds, that can't be fixed.

[real lynx]

What kind of grass are you looking for? Something pixelated? 2d Minecraft style?

Maybe you want something digital

I didn't like that it changed color

[marble cape]

well first this image cause its actually good: two things: first how do i see the full brown branch in that you see? and second how do make those herbs more like its planted on the ground you know?

and how do i make those herb more like an anime like yugioh trading card game artwork style cause i was picture being that kind of style

[real lynx]

I would erase the bottom part of the grass so it blends in better.

[marble cape]

oh i see

[real lynx]

I don't understand what he wants to do with the brown branch

[marble cape]

but what se the full view of this image i mean the brown branch and some of the herbs have been cut off

wanted to see the full image of that i could do that

[real lynx]

Sorry there are things I don't understand because I only speak Spanish and maybe the translator doesn't explain it very well.

[marble cape]

ohhh i see my apology

[real lynx]

I don't know the game, I'll investigate on Google

[marble cape]

or just like anime overall style

but how do i see the full image of that photo cause some parts have been cut off when i generated it

?

[real lynx]

You have to change the size

When I generated your image it was also cut

[real lynx]

I can't find the connection between your collectible card game and And its grass

[marble cape]

no its for this one for now

you see some of it been cut off

so i can see the full image

[dim zodiac]

My bad, didn't scroll back far enough, thought your ask was starting with "what's the best way..."

but the rest of what I said is true

I would switch to Sketch2Image mode, then use your background colour in the sketch tool, sketch over the text and prompt "background"

We don't really "talk" to canvas - it works most consistently to prompt it for what you want to see, just like in normal image generation

[marble cape]

So for the prompt just put in background that’s it?

Nothing else?

[drifting swallow]

I'm getting some errrors using the cavas, any tech issues at the moment?

[marble cape]

So for the prompt just put in background that’s it?

[dim zodiac]

sorry the advice there was for VallyWally

[marble cape]

Ohhh I see

But for the prompt you said just type in background for that image nothing else?

[manic root]

Hop into today's Daily Challenge and put your skills to the test! 🧠 Jump to #1108130218260041758 and give it a go! ✨

[drifting swallow]

Is there a bug-reporting/trouble-shooting area for the website?

[spiral oracle]

where do i type my negative prompts

[marble cape]

Can anyone show me? Those two things

[chrome kiln]

HELP: one of my long term pics I've wanted to make was a glowing white stag, with glowing pure white fur, no hint of blue, nor yellow, with a huge majestic 50 point antlers,

kind of like this, but not see through

[dim zodiac]

We did a "Patronus" series like this before, I wonder if I can find any of my old prompts...

[spiral oracle]

Does advanced settings support the 3D animation ai?

[real lynx]

Do you want to animate something in 3D or create something in 3D?

[spiral oracle]

I want to create like Pixar characters, animals that look like they're in Pixar films with costumes etc for t-shirts

[marble cape]

I’m still stuck

[real lynx]

Any character in particular?

Or just animals

[real lynx]

I see you having problems, encourage you to play with AI to learn

The best way to learn is by making mistakes

[spiral oracle]

IV been working on koalas a ton, I get the odd awesome one, but ofcourse I'm striving for more. Refining my prompts etc

For example

[real lynx]

It's very pretty

[spiral oracle]

I wanna throw them on a shirt and sell them 😁

Hopefully the print company can give top quality prints that are long lasting

[real lynx]

I think it's a very good idea

[spiral oracle]

If you're an expert id love to know some tips and tricks 😏

[real lynx]

A Pixar 3D animation style medium shot of a koala with a Mexican sombrero. It is sitting on a cactus in a desert landscape. The background contains a few more cacti and a mountain range.

If you want to add letters to your drawing you can do so in the Phoenix preset

XD

I wanted him in a suit but he got a doll

[spiral oracle]

Tsahahah check these ones

[real lynx]

I would dress up as superheroes.

[spiral oracle]

Oh I have look 😛

[chrome kiln]

ok @real lynx i got it

[spiral oracle]

@real lynx

[spiral oracle]

Adopt worthy?

that dude eats bad guys for breakfast

oki i think i have my prompt pretty sharp now

[simple robin]

I find Adobe Express webpage is very helpfull in scaling images more and removing background.

[hardy sentinel]

Of course adopt worthy, very cute

[red ridge]

Gift voucher template

[stark wedge]

Friends, I need to generate horror images, but the page doesn't allow it, it says something like "it's inappropriate" it's a real frustration and disappointment, but is there a way to generate content of that type that I need without that restriction?

[scenic arch]

We are building an app that generates illustrated stories using character reference. We are running into some challenges and I’m hoping someone might be able to help:

• since the image generation prompt is AI generated by GPT we keep getting an error saying that the prompt contains inappropriate things involving a child
• character reference quality isn’t always great, would fine tuning help? If so, any best practices for character consistency? Can I still change the image generation style to something different to my dataset while maintaining character integrity?

[real lynx]

Hi

You could ask for help with a command and a private room will be activated exclusively for you. The command is /mentoring

All mentors would be called upon to help you

[night vault]

1. programming wise, just have some filter list to remove words like child young boy girl, change them to appropriate age or prompt gpt to avoid such words. if you want child in prompt then you need to avoid sexuality and violence

2. character reference works better on real human face and any stylized cartoon may or may not work

[remote beacon]

How do you make sure text is not on an image?

[marble cape]

ok well how do i make this shield more of an anime style to it like sd gundman feeling

[marble cape]

ok so the closest thing that resembled that was this so far the color is off but somehow need to fix the coloring

sorry this

ill just go with this one

so yeah just if you guys can see to make this image wide for me to see the whole thing

[jade tinsel]

I like the style of an image I found in the library. I would like to try to copy the settings. What Preset is "Leonardo Style"?

[real lynx]

You can go into the image settings and tap the remix button to make it copy the whole style

[real lynx]

When the image goes beyond the edges and does not come out in its entirety, the size must be modified.

Unmodified size cropped the same as in the thumbnail

In configuration you have these sizes

[real lynx]

Regarding the colors you can modify them by adding things to the prpt.yndandpñeenos force to the image or simply changing presets

You can also modify the colors in canvas studio

[open echo]

And just to complete the answer 😄

"Leonardo Style" was when using "Legacy Mode". It was preset used while Alchemy is disabled to help add more contextual coloring and lighting to pictures without having to prompt for everything.

An exemple would be prompting for a witch in a medieval village. With Leonardo Style on, it might help get a moodier ambiance, gray sky, derelict building while leaving it at None might have been more litteral with what was prompted (and missing details)

With the advent of the new user interface and more preset options, it kinda got deprecated for better and more powerful features.

[marble cape]

let me see

where is this at?

This one doesnt look bad either!

its just a little process of learning

[marble cape]

ok it turned out like this maybe it needs dirt on the ground and also trying to implement the herbs on the ground still

[spiral oracle]

wddup gang

[proper crystal]

Leonardo keeps making me waste credits coz I want to change the colour of the text any way around it?

I have to keep my hand super still otherwise it'll end up wonky and uneven

[proper crystal]

Hello?

[dim zodiac]

Hey... What are you trying here, using sketch to image?

[proper crystal]

Yeah sketch to image

It takes forever to draw the text and if I draw it not straight it just looks wonky

[dim zodiac]

Another option would be to use the Edit with AI option in Phoenix

Assuming you made this in Phoenix, you can hit the Edit with AI button and say "turn the letter M blue"

[proper crystal]

How? The image is already generated it's too late

That makes no sense

[open echo]

There are two ways to look at it: you can either say « it makes no sense » or « a new feature I might not know about? Cool! »

[dim zodiac]

Thanks Tux! I was out swimming so just seeing this 😄

@proper crystal ^^^ this is exactly what I was describing, and it can ONLY be used on images that are already generated. Neat!

[proper crystal]

Yeah they've been already generated with phoenix but how

[thin tinsel]

or just press and hold shift.

[river pawn]

Hi all,

Like some other people commenting on here, I’m having some trouble with creating consistent characters. My goal is to create a comic strip, where I could create a small subset of base characters facing slightly left or slightly right in a variety of poses, with a variety of facial expressions. I’ve used over 2000 tokens experimenting with a number of different things, including using the character/style/reference features; locking to a specific seed; and training a model for a character. I feel like I’m making some progress but running into numerous issues that keep me getting to what I’m looking for, like a similar character being generated in different art styles whenever I change the prompt slightly or the character only facing forward. Ideally, what I’d like is to create a character and then prompt for that character to take on different poses in different orientations. Does anyone have any suggestions? Thanks!

[dim zodiac]

When you're looking at that image in the Generation View you'll see a little pencil with sparkles icon

Clicking that will open the dialogue box for Edit with AI

[simple robin]

someone said if you want to generate words in a picture, you need to put them between " " . But its not generating well.