There are other hogtied FLUX LoRA that work pretty decent, but...

If you try to create a hogtie in FLUX with the subject wearing heels or shoes, then it's very hard for FLUX to draw heels reversed upwards.

So, this is where this LoRA comes in.

It can create a human hogtied with the feet upwards and accurate rendering of shoes/boots with heels.

Plus. The model knows:

• coloured ropes (just add the texture/color you want)

• several types of gags: it's pretty good with ball gags (coloured ball gags and coloured leather strap). It can do cloth gags (coloured) and coloured cleave gags (although a bit thick ones).

• Tape gag (and some chance for wrapped)

You will see it tends to use boots (around 60% of the training images), but if you prompt it, it can draw any shoe with a heel. It is not that good at bare feet (maybe a next version would be with an addition of more barefoot hogties) and you will need luck to get something without heels.

Usage: Natural language. Prefer to place the hogtie description near the start of the prompt. A typical use case is: "A .. woman with [hair description] tied up in [colour] rope, lying on the floor [in ... setting], bound in a hogtied pose, with her high-heeled [boots/shoes] pointing upwards. ......". And then you can add whatever you want and even be more elaborate on the bonds.

Trained on a vast number of images (499), most of them with the subject wearing shoes or heels. All were captioned with joycaption and then corrected by hand. The epochs might seem high (30) but this is a very hard notion for FLUX (and I think it came out pretty well)