All Images Here Are Generated By AI.

All Models

>

PrintableHeroes TTRPG Paperminis + VTT Tokens

papermini, printableheroes, a misty foggy dark violet-blue silhouette of a shadow human knight, swirling churning circles of darkness making up his body, body made of shadows, shadow man, wispy tendrils of shadow mist streaming out from him, shadow, shadowy foggy dark violet-blue silhouette, standing on shiny obsidian ground, by Brom, Brom, (white background) <lora:vtt-tokens-v1-0-pruned:1.0>
papermini, printableheroes, a misty foggy dark violet-blue silhouette of a shadow human knight, swirling churning circles of darkness making up his body, body made of shadows, shadow man, wispy tendrils of shadow mist streaming out from him, shadow, shadowy foggy dark violet-blue silhouette, standing on shiny obsidian ground, by Brom, Brom, (white background) <lora:vtt-tokens-v1-0-pruned:1.0>

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to take AI photos of yourself, and more.

Model Description by Creator

PrintableHeroes TTRPG Paperminis

PERMISSIONS: As listed, but additionally provide credit to PrintableHeroes with a link to their website https://printableheroes.com/. I won't sue you. I don't really care. I'm a guy on the internet. Just have some respect for the artist whose style I'm ripping off.

This is a LoRA based on the PrintableHeroes paper miniatures intended for TTRPG game masters as an affordable alternative to expensive 3D miniatures. The PrintableHeroes paper miniatures have been a GAME changer for me as a game master and I highly recommend you pay just a few dollars on their Patreon for access to HUNDREDS of amazing print-ready paper miniatures. If it's a popular generic fantasy monster or it's in any of the D&D official monster books, there's probably already a good enough mini on their website. In the case that there is not a paper mini available, such as if you are homebrewing a creature, there is this LoRA! I know this sounds like an ad, but it genuinely changed my TTRPG combats (and saved my wallet).

This is my first attempt at making a LoRA, so it's bad. If you're good with LoRAs or pre-processing images and you want to help out, please contact me, and I will be glad to share the training data I have.

For obvious reasons, I can not provide the training images, but I did provide the captions to help the prompt engineers out there. If you really want to generate good images, it's good to have your prompts match up to the training captions.

For my first LoRA, it's kind of bad. It can't generate monsters without legs no matter how hard I try because I trained it on both humanoids and monsters, and it can't generate good looking humanoids. Overall, the results aren't great. In the future, I may release two different versions: one for humanoids and one for monsters. IMPROVED VERSION COMING SOON MAYBE™.

Right now, the LoRA is best at generating monsters. If you want to generate humanoids, you're wasting your time. Either get really lucky with really good settings or wait until I release the humanoid version of this LoRA.

TLDR: This LoRA makes 2D paper miniatures for TTRPGs, trained off of PrintableHeroes' paper minis (check them out, great resource!). Good with monsters, good luck with humanoids!


Prompting Guide

This prompting guide assumes you're using the AUTOMATIC1111 webui In this version, I made the mistake of not turning on caption shuffling. Whoops! Your prompts should follow the following structure:
papermini, printableheroes, a character portrait of a <your creature with important details>, <details, details, details>, <standing on|floating on|flying above> <description of ground>, ((white background)) <the lora:0.8>
Negative prompt: gradient background, gray background, colorful background, patterned background, sepia background, oversaturated, colorful

For the important creature details, include gender, species if it's a well-known D&D monster, and important details like what the monster is wearing/wielding.

For the negative prompt, "short" and "cropped" might help if you want a smaller monster. If you want a terrifying monster, add descriptions like "cute, adorable, innocent, small, soft, round, fluffy" to your negative prompts for truly hideous results.

Use whatever samplers and settings work for you. I use Euler A and it works for me, but YMMV.

Really, getting a good image isn't that hard, at least for monsters, because weird fingers, anatomy, blemishes, and general AI weirdness can actually make a monster look even more monstrous. It's not a bug, it's a feature!™

Printing It Out/Getting Your Monster on the Table

Once you've found your perfect image, it's time to get it onto your table.

I generated this image, which I sort of like. It's a Ratcrawler, a race of mutant creatures created as a result of the Darkplague who have overrun the dwarven mines of Vermilith. It's not perfect, but don't stress out. You can either spend an hour regenerating and running it through img2img and inpainting or worse -- manually touching it up -- or you can just ignore it because the blemishes won't really be noticeable as a 1/2 inch mini. If your monster is supposed to be larger than 1/2 inch (or a Medium creature in D&D), that's when you should put some effort into making it look decent -- because your players WILL notice. They'll be staring at that baby for hours while they argue over how best to kill your monster.

papermini, printableheroes, a [shark-like] rat monster with gray fur, long sharp terrifying teeth, large dirty mole digging claws, mole nose, pink rat tail, red eyes, long beaver front teeth, dorsal fin curving down its back, webbed paws, long rat-like tail, dirty messy rodent hair, rodent-like features, [shark-like features with sharp shark dorsal fin], amphibious creature, standing on rocky dirt ground, by Brom, white background <lora:papermini_v1-3:0.8>

Negative prompt: ((low quality, worst quality)), cute, adorable, happy

Steps: 20, Sampler: Euler a, CFG scale: 7, Seed: 1600317369, Size: 512x512, Model hash: cc6cb27103, Model: v1_v1-5-pruned-emaonly, Clip skip: 2

The easiest way I've found is to put my images onto some image host (there's a million of them. I use postimages.org) and create a printable sheet of them with MonsterForge, a really awesome open source website for making paper minis (contribute on GitHub if you can!). Go into the Quick Build page and add your creature. Give it a name if you want, add the image URL, configure the other options however you like (especially the size), and set the quantity. For now, I'll only need 20 Ratcrawlers.


Click "Generate my Minis!" and it'll download a PDF of your minis. Here's what my PDF looks like:

And, just like that, my Ratcrawlers are ready to be printed out. With some patience, scissors, and optionally glue, you'll have an army of Ratcrawlers to infest your mines with! Cheap, sort of easy, and customized to all your homebrewing needs!

If you don't like manually cutting these guys out, or you want nice silhouettes cut out instead of tented up rectangles, you can buy a vinyl cutter. I use a Silhouette Portrait 3, but anything else will work. However, that's way out of the scope of this tutorial. If you're familiar with their software, it shouldn't be too difficult.

Show me some of your creatures! I might want to "inspire" them into my campaign ;)


Getting VTT Tokens on the Tabletop

Running a virtual tabletop? No problem! You can generate VTT tokens.

First, you'll need to generate a good looking VTT token. I can't help you with that, because I haven't experimented much with the VTT LoRA, but it generates fairly good results without too much haggling -- just crank your batch count up, set a random seed, or even use X/Y/Z grids and go wild.

Once you've got your image, you'll need to clean it up and get it ready for the VTT. You can use any image editor for this.

Here's the image I generated. It needs some cleaning up, but otherwise it looks great!

The image isn't perfect, and there's a few glaring artifacts. You should either use img2img if you're confident in that or just fix it up in your image editing software. For this tutorial, I'll ignore the artifacts.

Using the magic wand tool, I removed the white background.

Now you've got your token PNG, and you can toss it into your VTT.


Alternative Method

If you find the VTT token model doesn't generate good results, you can use an image generated by the basic model, upscale it, toss it into an image editing software, crop it, place a circle around it, and replace the white background with some nice gradient.

I went from this image:

...to this:

All using 2 minutes of image editing. Nice! This alternative method is cleaner, but it requires more manual labor.

Try Full Demo
Quick Generate
Image quality might not be as good as what you can achieve with our full generation tool.
Placeholder
Choose A Style, Type What You Want In The Image And Click Generate
Placeholder
Choose A Style, Type What You Want In The Image And Click Generate
What you want in the image.
Clear
Random
* This model is available for use and will generate similar character to the ones seen in the images.
Train Your Own Model

Images Generated With This Model

Loading...

Generate AI Images. Simple and Easy

AIEasyPic is the #1 AI image generator. Generate images with simple prompts, swap faces, train models to reimagine yourself, and more.