Give me a prompt and i'll make an image

[trappedslider]

perhaps you'll do better than I did with prompt: "The Hindenburg colliding with the Titanic"

This was done https://creator.nightcafe.studio/

 

[Bohandas]

This was done https://creator.nightcafe.studio/

That's WAY better than what I managed on Artbreeder (while still falling short of the image in my head of a downward tilted hindenburg directly impacting the ship head-on; AI seems to have trouble with the colliding part, or at least the models I tried all did and apparently this one too. I also couldn't get it to create a pile of dead eagles - they were always soaring majestically even if I put 'majestic' in the negative prompt)

 

[trappedslider]

That's WAY better than what I managed on Artbreeder (while still falling short of the image in my head of a downward tilted hindenburg directly impacting the ship head-on; AI seems to have trouble with the colliding part, or at least the models I tried all did and apparently this one too. I also couldn't get it to create a pile of dead eagles - they were always soaring majestically even if I put 'majestic' in the negative prompt)

it helps to if you tell what is in your mind,they're not dead

If i really invested the time and effort into it I could more than likely get to make what you're thinking of.

 

[trappedslider]

A fierce female warrior stands confidently in an industrial, dystopian environment engulfed in chaos. She wears a sleek, tactical black jacket with multiple pockets and straps, exuding a commanding presence. Her blonde hair is tousled, framing her determined blue eyes. In one hand, she holds a futuristic weapon with a firm grip, ready for action. Behind her, fiery explosions light up the industrial background, filled with metal beams, pipes, and scattered debris. Smoke billows around her, enhancing the dramatic, action-packed atmosphere. The lighting is dynamic, casting intense shadows and highlighting her bold features and powerful stance.

 

[trappedslider]

 

[Bagpuss]

That's WAY better than what I managed on Artbreeder (while still falling short of the image in my head of a downward tilted hindenburg directly impacting the ship head-on; AI seems to have trouble with the colliding part, or at least the models I tried all did and apparently this one too. I also couldn't get it to create a pile of dead eagles - they were always soaring majestically even if I put 'majestic' in the negative prompt)

AI art programs struggle to create anything they have never been trained on. A prime example of this is ask it to draw a glass of red wine full to the brim. It can't as it has only ever seen wine glasses full to halfway. Similarly it won't have images of collision like the one you want or dead eagles.

 

[trappedslider]

So, I decided to look into the pony fork of stable diffusion, it's prompting is all done on Danbooru tags like 1girl,witich etc

This was done with: score_9, score_8_up, score_7_up,1girl,witch,fantasy,casting spell,rooftop,rating_safe,

 

[trappedslider]

this was done using a pony version that was trained on realism

 

[trappedslider]

I think i'm just going to keep with flux,it's simpler for me to do prompts for

 

[trappedslider]

Some more work with pony