trained on LTX 2.3 full dev model
This lora is trained on sheer clothing and dresses.
twerking
and panties and general other small titbits. - motion etc.
uploading it as part of my workflow (the twerking video) - that was in it.
didn't want to make it so people couldn't recreate what i show.
it was 200 video lora for 7k steps probably train a little more tonight <3 (not sure it needs it )
Description
FAQ
Comments (6)
High class! You aspire to greatness. You bring out the best of us.
Thanks bro!
Appreciate you!
Able to share any more about the vids you trained on? What resolution and how many frames?
I was able to train an LTX 2 character lora using AI Toolkit locally on my 4090. My next goal is to train a motion and/or IC lora. Pretty sure I'll need to run on cloud (runpod likely for me).
What software+config are you using to train?
Thanks for posting. Any particular trigger words we should be using? I'm guessing "twerk/twerking" and "sheer/see-through" but let us know if there's anything else.
Great work! I2V or T2V?
everything installed, nut this error
Failed to validate prompt for output 140:
* LTX2PromptArchitectQwen 464:
- Value not in list: 🌍 environment: '1529394688' not in (list of length 59)
- Value not in list: 💋 emotional state: '192' not in (list of length 33)
- Value not in list: 🎵 music genre: 'False' not in (list of length 51)
- Value not in list: 🤖 model: '' not in ['huihui-ai/Huihui-Qwen3.5-9B-abliterated', 'huihui-ai/Huihui-Qwen3.5-27B-abliterated', 'huihui-ai/Huihui-Qwen3.5-35B-A3B-abliterated', 'huihui-ai/Huihui-Qwen3.5-35B-A3B-Claude-4.6-Opus-abliterated']
- Failed to convert an input value to a INT value: 👥 subject count, , invalid literal for int() with base 10: ''
- Value not in list: control_after_generate: '0' not in ['randomize', 'fixed', 'increment', 'decrement']
Output will be ignored