Missionary Penetration (Side View)

LORA
Reprint


Updated:

No showcase images available, the model won't be visible to others.

14K

All credit to jacob42

Missionary sex position with penetration. Trained on various images ~1024 that include penetration in the missionary position. Trained off the RealXL model using ~3200 steps (80 per image, ~20 photos, 2 epochs). WIP

Generation tips: Include "penet" as your first tag! Simple prompts like "naked woman and man having sex on bed" are great. You can also specific "missionary" or "penis in vagina" etc. to drive the point in. But often not necessary. It's best to modify the strength of the Lora and the denoising strength. To start, using 0.8-1 for the Lora strength is good to get the model. Then, I use img2img for the best results. If you have a good pose but bad anatomy, using a few generations of img2img (playing with denoising strenth, I use 0.2-0.8 depending on what I need), you can get much clearer images. If you turn down the Lora strength during this stage (maybe 0.3 or so), you can get characters looking more like RealXL ones. Then, if I want to change something small like a face or remove an extra limb, I inpaint a mask. If I want to preserve the penetration but generate completely new people, I inpaint mask on penetration but select everything not in the mask to be regenerated. This is good for preserving penetration with an inpaint mask and then regenerating the people with a lower Lora strength. Results may vary. Play around with the numbers! Img2img, inpainting, denoising strength, and Lora strength are your best friends.

Version Detail

SDXL 1.0
All credit to jacob42 Missionary sex position with penetration. Trained on various images ~1024 that include penetration in the missionary position. Trained off the RealXL model using ~3200 steps (80 per image, ~20 photos, 2 epochs). WIP Generation tips: Include "penet" as your first tag! Simple prompts like "naked woman and man having sex on bed" are great. You can also specific "missionary" or "penis in vagina" etc. to drive the point in. But often not necessary. It's best to modify the strength of the Lora and the denoising strength. To start, using 0.8-1 for the Lora strength is good to get the model. Then, I use img2img for the best results. If you have a good pose but bad anatomy, using a few generations of img2img (playing with denoising strenth, I use 0.2-0.8 depending on what I need), you can get much clearer images. If you turn down the Lora strength during this stage (maybe 0.3 or so), you can get characters looking more like RealXL ones. Then, if I want to change something small like a face or remove an extra limb, I inpaint a mask. If I want to preserve the penetration but generate completely new people, I inpaint mask on penetration but select everything not in the mask to be regenerated. This is good for preserving penetration with an inpaint mask and then regenerating the people with a lower Lora strength. Results may vary. Play around with the numbers! Img2img, inpainting, denoising strength, and Lora strength are your best friends.

Project Permissions

Model reprinted from : https://civitai.com/models/127649/missionary-penetration-side-view

Reprinted models are for communication and learning purposes only, not for commercial use. Original authors can contact us to transfer the models through our Discord channel --- #claim-models.

Related Posts

Describe the image you want to generate, then press Enter to send.