New release (1/15/26):

I think I achieved a decent balance on the quality of T2V, I2V, and audio so I'm releasing this as a beta. Some times things go weird. Lower strength can help sometimes with trickier prompts. I really like the use of ltx-2-ic-detailer-lora with this lora.

I'm still working on my workflow but currently I'm running a video/audio training cycle then and image training cycle to improve genitals.

Differences from v0.1

Improved audio,
T2V - Improved penis (still not perfect, but way better)
I2V - Similar or better results

Tags used during training

A woman is lying on her stomach in prone position a man behind her thrusts his hip forward and back sliding in and out.
The mans penis is visible.

Audio tags

clapping cheeks
moans, moaning, the woman's breathless moaning
heavy breathing

Training Details v0.2

30 dataset videos 576x1024@121f and 1024x576@121f
30 high quality images 1024x1024
Frame Rate: 25fps
Steps Video: 4000 (Video was trained faster than audio)
Steps Images: 3800 (Used to improve penis appearance)
NO abliterated used

Generation details:

Workflows in all images in the showcase for release.
No abliterated model used. (just don't user the LTX prompt enhancer.)
T2V vids are fp-8-distill
I2V vids are 19b-dev full.

Prone Face Cam - v0.2