D&D top down token
Resolution
Model was trained on 512x512 images, so it works best on this resolution.
Prompting
Model was trained on D&D monsters, heroes and master characters (NPCs). So it works best for medieval fantasy setting. It's recommended to use trigger words, that define monster type.
But it's not overtrained only on this setting, as for me works pretty good with almost everything I tried. You can not use trigger words at all, that would be fine too, sometimes that's even better.
Also you can try mix them to get interesting results. Check all example images for inspiration.
Link to huggingface: https://huggingface.co/Zapper/top-down-token-v1-0
Description
FAQ
Comments (11)
Looks awesome
Nice! This looks a lot like Caeora's style, was it trained on their tokens? Either way, will be trying this later!
Does not work. Need a .yaml config file to work.
Loaded config file, but it's default 2.1 512
@Zapper Wanted to request a SD1.5 version to work with ComfyUI and such, but it is somewhat ahead of its time. Any ways of doing sometime similar with scenery? (maybe a LoRA?)
Doesnt work
Loaded config file. Try to put it to your model folder.
@Zapper It works now, thank you.
Would love to see a revival of this trained for SDXL.
Well, I thought about that but training pics are around 480x480 and not sure how this will look like extended to 1024x1024
Also my PC is not powerful enough for checkpoint training, but maybe I'll try Lora. Or will try to rent gpu.
It doesn't work for me, I put the checkpoint and the yaml file in my template folder but all my creations are in a kind of haze
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.