Could/Would one or five or you so good at doing this start up a thread with a short primer or links to such on how one could get started with this? I'd love to be able to do a few images here and there for my home games. But don't even know where to start.
Honestly, the easier way so far is to rely on OpenAI's Dall-E 3.
It is available "unlimited" for ChatGPT subscribers, but there is a free demo through Bing Image Creator :
Intelligent search from Bing makes it easier to quickly find what you’re looking for and rewards you.
www.bing.com
It requires a Microsoft account, which is free. It allows you to make a few generations with free credits that replenish over time (I don't have kept current with the rate). It's the easiest tool available, just type a description of the image you need, and it will give you four samples, based on what you asked for.
The pay version allows you to iterate around an image, but it's not very precise, it will generate another randomized image, just using the elements your asked in the prompt of top of the initial prompt.
There are other tools that allows for more customization of your images, like Midjourney, but they are paid for.
The most customization you'll get is by using a tool called Stable Diffusion, by Stability AI. It is, however much more complex to use, requiring installation of software on your computer and its performance is dependant on your graphic card (not because it's graphical, but because all AI works need a very fast memory, that is available on GPUs only). It is the best and most economical way but it requires some effort to install. If you want to create a few images, it might be overkill.
You can try some of them on these free sites:
Pricing - Playground <= there is a free mode that let you generate images daily using their custom model
https://creator.nightcafe.studio/ <= technically not free, but you get free credit daily by clicking on an email link.
Easy Diffusion Online, Easy Stable Diffusion, FREE AI Image Generator. Fine-tune image generation with checkpoint and lora selection.
www.easydiffusion.online
You can choose among some models, the size of the image, the prompt (description of your image) and negative prompt (words listing things you don't want to see in the generated image), but you can't accesss the shiny tools that I spoke of earlier.
If you're interested :
https://stable-diffusion-art.com/install-windows/ has a tutorial on installing. But mastering the tools take time (and more of them are published each week...)