Dungeons and Diffusion v3

New version 3 is trained from the pre-eminent Protogen3.4. I found that training from the photorealistic model gave results closer to what I wanted than the anime model. This model is fantastic for discovering your characters, and it was fine-tuned to learn the D&D races that aren't in stock SD. Works best with 768x512 and other resolutions larger than 512x512

Try a prompt like:

Positive: modelshoot style digital illustration medium shot of an [older woman] elf wearing a long patterned (kilt) and [breastplate], holding a partisan spear and a shield, wearing an (embroidered) traveling quilted cloak, with white hair up in a bun, art by tony sart, wlop and rossdraws and Magali Villeneuve, sharp focus, intricate background of a countryside, hills, farmland, farm building, trees, trending on artstation, trending on cgsociety, rustic, fine art, vivid colors, detailed face and eyes [Brenda Fricker | Katie McGrath | Saoirse Ronan], [flowers and lavender and herbs in belt], [cape], stars, galaxies, nebula, runic symbols, wooden earrings, [samurai], purple gravity energy aura