Give me a prompt and i'll make an image

[LuisCarlos17f]

I was asking the first one. Sword & Sandal is a cinematographic genre about stores in the ancient age, before the fall of Rome.

 

[Mad_Jack]

Indeed Swords, and Sandals.

Now we need a character wearing sword sandals.

 

[trappedslider]

I was asking the first one. Sword & Sandal is a cinematographic genre about stores in the ancient age, before the fall of Rome.

That's what i figured you were wanting, but I figured It would be helpful to show what happens when it's really vague lol or when the AI takes it literal lol

 

[trappedslider]

Now we need a character wearing sword sandals.

 

[Blue]

I've got a Stable Diffusion version installed on my PC, but I've barely got enough VRAM (6GB) to do things at a reasonable size. One thing I have been playing around with is a lot of different models which can get you rather drastically different results.

One of the discussion points that comes up every once in a while is how "horny" the model is. Because one of the reason people run it locally is to have an implementation that disables the SFW filters. So lots of the models out there are trained including NSFW images. And the more that models will include risque elements where the prompt is silent is how "horny" that model is. I often have "sfw" as one of my prompts and "nsfw" in my negative prompt.

Another big point about the models is tagging. There is a large anime-style repository called Danbooru that was used to train a model called Waifu Diffusion, which then has gotten mixed into many other ones. But Danbooru has in-depth tagging done by people for years and years and are well documented, so it kept those tags. But those are different tags than used elsewhere, so prompting can get odd - especially when mixing Danbooru and non-Danbooru models.

 

[Blue]

"Was it a dream where you were where you see yourself standing in sort of sun-god robes on a pyramid, with a thousand naked women screaming and throwing little pickles at you?"

Ah, Val Kilmer from Real Genius. Fun movie.

EDIT: Oops, responded before I saw the clip. Classic quote tho.

 

[Rabulias]

I briefly tried a couple of the free AI art generators and had little success with working up an illustration of a villain for my campaign. Might be my inexperience with writing proper prompts, and/or I think the generators I tried do not recognize some D&D terms. So I appreciate any help you can offer!

A grim-faced, middle-aged, male derro, wearing black leather scale armor and wielding a rapier. He is clean shaven, but has a full head of slightly wild hair. He has an almost vertical scar starting above his left eye, continuing below it. His left eye is bluer/milkier than his right eye. If you can give him a smirking sneer, all the better!

Thanks for any images or prompt-writing advice!

 

[trappedslider]

I've got a Stable Diffusion version installed on my PC, but I've barely got enough VRAM (6GB) to do things at a reasonable size. One thing I have been playing around with is a lot of different models which can get you rather drastically different results.

One of the discussion points that comes up every once in a while is how "horny" the model is. Because one of the reason people run it locally is to have an implementation that disables the SFW filters. So lots of the models out there are trained including NSFW images. And the more that models will include risque elements where the prompt is silent is how "horny" that model is. I often have "sfw" as one of my prompts and "nsfw" in my negative prompt.

yeah, like this one

Spoiler

, and also the AI sometimes outright ignores everything but one word. I have I think 10 different model checkpoints and a few embeds. The other thing I'm learning from reading prompts at https://prompthero.com/ is giving more weight to which words or groups of words.

 

[trappedslider]

I briefly tried a couple of the free AI art generators and had little success with working up an illustration of a villain for my campaign. Might be my inexperience with writing proper prompts, and/or I think the generators I tried do not recognize some D&D terms. So I appreciate any help you can offer!

A grim-faced, middle-aged, male derro, wearing black leather scale armor and wielding a rapier. He is clean shaven, but has a full head of slightly wild hair. He has an almost vertical scar starting above his left eye, continuing below it. His left eye is bluer/milkier than his right eye. If you can give him a smirking sneer, all the better!

Thanks for any images or prompt-writing advice!

Unless, it's been trained on D&D, getting a species that have only been in D&D will be tricky.
But here's my best attempt with your words

Spoiler

 

[Rabulias]

Yeah, that's better than what I got, but it's still far from what I have in my mind (And definitely not a derro. Does anyone know of a "D&D trained" art AI?). Oh, I endure the curse of having no artistic talent! If nothing else, my experience shows there will still be a place for human artists to cover things that AI does not grok.

Thanks again!

 

[MarkB]

Yeah, that's better than what I got, but it's still far from what I have in my mind (And definitely not a derro. Does anyone know of a "D&D trained" art AI?). Oh, I endure the curse of having no artistic talent! If nothing else, my experience shows there will still be a place for human artists to cover things that AI does not grok.

Thanks again!

Maybe just use prompts based on physical description? Like, grey skin, white eyes, white hair.

That, or spend 20 minutes in Photoshop with it afterwards.

 

[trappedslider]

Yeah, that's better than what I got, but it's still far from what I have in my mind (And definitely not a derro. Does anyone know of a "D&D trained" art AI?). Oh, I endure the curse of having no artistic talent! If nothing else, my experience shows there will still be a place for human artists to cover things that AI does not grok.

Thanks again!

Turns out there is a checkpoint model trained on D&D Dungeons and Diffusion v3 | Stable Diffusion Checkpoint | Civitai I'll download it later and try again

 

[Rabulias]

Maybe just use prompts based on physical description? Like, grey skin, white eyes, white hair.

I had not thought of that! Good idea!

That, or spend 20 minutes in Photoshop with it afterwards.

Yeah, with my "talent," that might make the picture worse...

 

[trappedslider]

Yeah, that's better than what I got, but it's still far from what I have in my mind (And definitely not a derro. Does anyone know of a "D&D trained" art AI?). Oh, I endure the curse of having no artistic talent! If nothing else, my experience shows there will still be a place for human artists to cover things that AI does not grok.

Thanks again!

How's this?

Spoiler

 

[Rabulias]

That is getting closer to looking like a derro; the D&D training helps. I will have to test out the idea that @MarkB suggested with detailed physical description. Thanks again!

 

[LuisCarlos17f]

Now I am thinking about a blue-skin female goblin, but cute, style waifu-monster girl. The blues are a goblin subrace what appeared in Expanded Psionic powers.

 

[trappedslider]

Now I am thinking about a blue-skin female goblin, but cute, style waifu-monster girl. The blues are a goblin subrace what appeared in Expanded Psionic powers.

how's this?

Spoiler

['/spoiler]

 

[Mad_Jack]

That is getting closer to looking like a derro; the D&D training helps. I will have to test out the idea that @MarkB suggested with detailed physical description. Thanks again!

If it helps, I pasted this together...

 

[CleverNickName]

Young woman laughing with a salad.

Spoiler

I've got to be doing it wrong.

 

[Whizbang Dustyboots]

I think we now know where the idea for Smile came from.