Better face consistency in V2. Well done!

v2 is so refined it's like night and day!

This LoRA is fantastic. Thank you.
Great face, clothing, items and background consistency.
And it is nicely customizable without much issues, as much as you stay in the same thematic.

I didn't have the time to combine it with other LoRA though.

Please, do other sex positions with the same quality !

Second version is perfect. Very flexible. Only one minus - woman have a small unproportional feet sometimes

v1 is great! I can't wait to test v2

I like the first version better, I'll use it for now.

This work is absolutely fantastic!!

Love this Lora, has anyone found a reliable way to use the male in a reference image as the man having sex with the woman in the reference image?

On Wan 2.1 with fusion x the LOW there is no way to make it work, it is as if it is not perceived at all, it does not give any hint of functioning despite the trigger.

Simple and effective. Keep up the good work! Here's a buzz to show my support.

Works surprisingly well. This face consistency LORA does help (not perfect): https://civitai.com/models/2119634?modelVersionId=2397788

This is really good I like how it’ll sometimes create fanciful things for her to throw her knee up on.

Still getting some extra limbs. Face seems better.

More and more people are creating Lora scene transitions, and I like its simplicity and the ease of writing prompts.

how do you make this lora is it just a person standing there then edit a sex scene and use only 1 video?

Another, "This is a really good effect." The point of weakness I see is that the penis tends to be very small and positioned too low on the male's body. Honestly, kudos on getting the penis and vagina to work at all. Will have to try some other LoRA's for more support there.

When using i2v, I can’t seem to control the character’s face at all. The character keeps turning toward the camera, smiling, or looking like they're talking. I’ve tried using negative prompts but it didn’t help. I’m not sure if the issue is from my workflow or the gguf checkpoint. Does anyone have tips or a solution for this?

Any ideas on why the output seems like doesn't have enough steps? Extremely blurry and washed out. I've tried with my known good workflows as well as workflows from sample videos. I can't figure out what is missing.

use this prompt. can gen dog gy style.

The screen goes dark and then lights up again to show side View the china woman having s ex with a man. reverse_cowgirl_lie_front_va gina. Both the fat man and woman simultaneously turn to look at the camera. The woman's face is shown throughout the entire video.

anyone know any other lora's like this that moves to the correct position instantly for other sex poses? I find that this is the only way I can insert a consistent male character fucking a consistent female character using i2v, since it doesn't have to bother trying to actually physically maneuver to get into the position.

Please tell me what the dataset looks like, I want to try training a similar LoRa . Please tell me how to correctly create a dataset for similar LoRa.

broo it's awesome, more position please, anal probably

banger ! did you make more ? cant find any