This lora is focused on intimate moments captured from front facing camera ideal for subjects facing the camera in doggystyle, plow, prone, or ¾ front positions, it can work from the side but no recommended. This LoRA's aim is to create accurate dynamic movement, as well more facial expressions:
Keywords, phrases:
→ Short thrusts, fast thrusts, slides in and out, penetrates
→ Fast, slow, rough
→ Pulls her hair, hands on neck
Workflow:
Here is a quickly organized and noted WF video. Save it and drag it into Comfy UI. https://civarchive.com/posts/25545461
Initial images:
For the V2.1: Starting images are a mixture of Z-image base checkpoint, and Pony CyberRealistic V15.0. Then I add do some enhancements with Zimage, Qwen, Qwen-edit to make them pop a little more.
The dark 'Raw' styled images are z-image and a style lora I'm working on.
Personal Settings:
Workflow posted above now, leaving this here for quick reference.
Settings for all V2.1 videos:
3 Advanced Ksamplers 8 step total, shift 5:
High noise, 2 steps, cfg:3 Euler, simple, add_noise: enable, return with noise: true, Without lightning lora.
High noise, 2 steps, cfg:1 Euler, simple, add_noise: disable, return with noise: true, High noise lightning lora.
Low noise, 4 steps, cfg:1 Euler, simple, add_noise: disable, return with noise: disable, Low noise lightning lora.
Description
Corrects over trained 2.0 version.
Adds penetrates as a keyword.
More training data with implied penetration to correct movement.
Details
Files
sfbehind_v2.1_low_noise.safetensors
Mirrors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors
sfbehind_v2.1_low_noise.safetensors