General Reminder: Each of these versions is a completely different model - check version info for usage instructions and example images as a baseline.Does pretty much what it says on the tin - this is a collection of LoRAs that produce images consistent with giantess foot fetish. As I've trained more, "giantess feet" is a bit of a broad umbrella - more specialized models will likely be put in their own LoRA pages. This will remain a collection mostly focused on people under or between giantess feet. Generally tested so that weight 1.0 is optimal (though tacking on other LoRAs usually requires you reduce the weight, as do some checkpoints that are less amenable to LoRAs), generally supports anime and toon-style checkpoints.UPDATE: I've been in the process of retraining a bunch of these for illustrious, as I find that it is far more stylistically stable and prompt adhering than Pony 6. Any of my illustrious versions are a significant improvement over the Pony versions, so much so that the older ones might as well be deprecated (not Pony's fault necessarily - I'm also better at this than I was when I trained those, but it is why I'm not bothering to retrain those at least until there's an improved Pony model). There is also a decent amount of cross-compatibility between illustrious and pony LoRAs, so if you have a favorite pony checkpoint, you're probably better off trying the illustrious version instead.Completed:Underfoot POV: Images in which the viewer is beneath a giantess' feetUF_tiny: Images in which a tiny person is under a giantess' feet (size scale - tiny person roughly the size of her foot or smaller)UF_mini: Same, but with more of an 'amazon' size scale where the tiny person is less than half the giantess' height, but larger than her footImminent stomp (illustrious): Images in which a tiny person is under a giantess' foot, usually with the implication that she is about to step on them.Between feet (illustrious): Images in which a tiny person is pressed between a giantess' feet.In progress:between toes: self-explanatory - will see how well this comes outunder toes: same as aboveSimilar models:<a target="_blank" rel="ugc" href="https://civarchive.com/models/1305822/giantess-feet-bondage">GTS feet bondage</a>: Images in which a tiny person (or the viewer) is bound to a giant woman's foot. Tied, leashed, or taped.Sock/shoe entrapment (WIP): This is what I actually started out wanting to do, but it proved trickier than I expected. I've revamped some of these datasets and will start retraining once the last two versions for this model are done.GTS Foot worship (WIP): I've attempted models like this and had trouble making something that works. It's tricky because it's nearly impossible to prompt poses for two different subjects simultaneously. The tiny subject is also often at the low end of what can be resolved without upscaling and careful inpainting. I think the approach is to create something where the prompt mostly poses the giantess and the model just knows what the worshipper should do, but that requires some very granular training of specialized models.Deprecatedunder_foot_mini (pony), under_foot_tiny, under_foot, gtsFeetPOV, : just use the illustrious versions above.imminent stomp, between feet (pony): hopefully the illustrious version is out soon2.1,2.0,1.0: Don't use these. Ever. I didn't know what I was doing.

Giantess Feet

Take 2 at this LoRA. Far more consistent results, and much better about putting people in the right place. Epoch 10 has a bit of same-face syndrome and I'm going to tinker with styles and characters. It might be a bit better about foot anatomy, but it's hard to tell. Epoch 6 seems a bit more flexible with its style and characters, so I'm putting it here for now. I'm going to see how Epoch 10 responds to style prompts and add that one as a subsequent version if it works well.gtsFoot is the real trigger - the others were things I tried to push it to recognize. I suspect that I really need to independently train those poses and concepts and merge them if I want those to work, but the LoRA itself should give a grab bag of such concepts.

Files with SHA256: 7e4c7a069d970b14326dc02ab9681a25c9ceb79899b8caf3735c8d30e994189d