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.