You can support me here to be able to render better models (https://ko-fi.com/djo226). Supporters can request model.

Exclusive models are also available on my Patreon.

Prompt : AVTR, realistic, natural lighting, post-processing, ultra realistic, highly detailed, very detailed, cinematic, hyper realistic, intricate scenery, sharp, photo by Annie Leibovitz, film, studio lighting, detailed skin, ultra realistic, bokeh, sharp features <lora:AVTR:0.8>

Negative prompt: (deformed iris, deformed pupils, semi-realistic, cgi, 3d, render, sketch, cartoon, drawing, anime, mutated hands and fingers:1.4), (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, disconnected limbs, mutation, mutated, ugly, disgusting, amputation

Steps: 90, Sampler: DPM++ 2M Karras, CFG scale: 12, Seed: -1, Face restoration: GFPGAN, Size: 768x1024, Model hash: 81086e2b3f, Model: realisticVisionV20_v20

Face restoration is optional, may provide better result if not activated.

Use AVTR as trigger.

New process to get better result :

Set a prompt with a low weight (like <lora:AVTR:0.4>) and the action/situation you want (even with a lower resolution like 512x768, it'll be faster and avoid duplicate). Send the result of your choice to img2img, set the weight to <lora:AVTR:0.8>, set a higher resolution but keep proportion. Enjoy!