The quality of your generations may be heavily improved by using textual embeddings in your negative prompt. The embeddings I use are listed below and in the recommended resources.
The style and sampler I used for preview images is listed below:
Sampler - DPM++ 2M Karras
Positive - (masterpiece, best quality, 4k, detailed, intricate, realistic),{prompt}
Negative - (worst quality, low quality:1.4),( FastNegativeV2,easynegative, negative_hand-neg),{prompt}
Images generated using Berry's Mix VAE and 4X AnimeSharp Upscaler at Clip Skip 2
My typical settings for images:
Sampling Steps - 35 to 40
HiRex Fix - Upscaled by 2 at 15 to 20 steps, 0.35 Denoising Strength
Dimensions - 512x512, 512x768, or 768x512
CFG Scale - 5.5 to 7
One pass of ADetailer HIGHLY RECOMMENDED
I typically specify eye color/expressions/facial features here (blue eyes, grin, smug, mole, freckles etc).
This specific ADetailer model is made to detect different types of faces, so you may need to crank the detection confidence up or it'll start finding faces in clouds, trees, or whatever.
See images for examples of how I prompt. I prefer short, simple, tag-based prompts, but there is probably a lot of room for flexibility/improvement here.
I'm not yet completely familiar with more realistic models, so you may be able to modify the prompting to get better results.
Merging weights and process are available in the 'About this version' section per version.
All credit goes to the original model authors, please check them out.