cinematic photo dark evil lighting, [dark orange red scaled argonian lizard warrior]:1.3 wearing [dented battered silver chanmail armor]:2 holding [spear]:1.2 . 35mm photograph, film, bokeh, professional, 4k, highly detailed
Negative prompt: drawing, painting, crayon, sketch, graphite, impressionist, noisy, blurry, soft, deformed, ugly
Steps: 120, Sampler: DPM++ 3M SDE Exponential, CFG scale: 4.5, Seed: 2349427416, Size: 512x768, Model hash: f61328e842, Model: RealitiesEdgeXL_4, VAE hash: 235745af8d, VAE: sdxl_vae.safetensors, Style Selector Enabled: True, Style Selector Randomize: False, Style Selector Style: Photographic, Version: 1.6.0
Now use ChatGPT for the stats.
Impressive! I always had problems generating non human characters, recently I generated a cool looking white kenuk from my campaign, maybe I will share it on this community.
About a year ago, I was thinking about creating a little app as a personal project, that would combine stable diffusion with these random character generations, and then spit out an image like you did. Never got any further than figuring out how to query the SD API using javascript.
This is the site I wanted to use for character prompts: https://whothefuckismydndcharacter.com/
I’m impressed you got the weapon right. I rarely have a weapon that isn’t floating, or just completely through the head of the subject. Nicely done.
Isn’t that the one chick from The Wheel of Time on Amazon?
