Lora trained on 168 images of feet facing viewer while lying.
I release all my models for free, if you feel like supporting me and my work you can support me here and buy me a drink.
Toes are kinda random, chance is somewhat slim...
Prompts:
<lyco:feetaway1-000006:1.0>, on bed, indoors, simple background, feet, ass,
Trained on NAI, gen'd on AOM2 hard.
Description
FAQ
Comments (24)
Lmao these feet loras are kinda random...
I've notice that if you reseed from width on seed variation with most of these kinda models, you can come pretty close to getting a proper number of toes. Other than that, idk why AI is so recursive with toes and finger lmao, the technology is ahead of my understanding.
I love some feet lora but most have been pretty miss. I'll give this one a shot! I'm trying to get consistent kneeling, view from back, with feet soles. Basically can see feet, side boob, butt, and side of face. The current one on civitai gives me chunky toes. Any chance you would be able to do a lora like that? I'll try using this lora with "kneeling" prompt and see what I get too.
@servbot yeah this one is just laying down view image sets. I know what you mean tho Ill add it to the list
@servbot there's a lora called "true kneeling" or something, might wanna check it out
@xyskl yeah thats the one that gives me chunky toes
@servbot you could try inpainting them without the lora
Is it LoRA or LyCORIS - in your prompts you use <lyco:feetaway1-000006:1.0> but Civit and your description seems to present it as LoRA?
Its a lora. I use the lycoris extension which allows all loras and lycoris files to be placed in the lycoris folder it all just works.
Ok thanks 😊One day may come where we'll be able to use both of them interchageably without always pay attention to that detail.
@Gaspatcher https://github.com/KohakuBlueleaf/a1111-sd-webui-lycoris I just disable the locon extension and only use the lycoris folder for everything
There is a special type of Lycoris on the civit. Could you please change to this type so that it will filter correctly on the site
@chimisuke this is a lora
do you mind giving me the date for this, i wanted to do a feet lora and would like to see what you did for yours
I think its just kneeling + feet prompt images
@Numeratic kneeling + feet prompt images? do you mind explaining,
@isaac04murray930 just gelbooru tags
@Numeratic i wanted the date, not the images. i already have a bunch of images ready, the lora works really well with my model i wanted to know the tags you used
@isaac04murray930 i prune out tags so there arent any besides the regular ones describing the scene / character. The pose prompts arent in the dataset
@Numeratic do you mind telling me how you used a custom model for training? i says to put in the url but when i did that it gave me a keyerror
@isaac04murray930 i use the basic Novel AI leak for training. I dont use collab for training so I have no idea what errors.
Does it work with realistic models?
No idea... only trained on anime
Hello, Im not sure what I am doing wrong but none of my generations look the same as yours, I am basically using the same model abyssorangemix hardcore, 2m Karras Sampling and my prompts are: on bed, indoors, simple background, feet, ass, <lora:yaemiko1:0.8> <lora:feetaway1:0.8> and some negative textual inversion ones. The feet I get are very blurry and the pictures are very bad quality every time. Can you give me some tips? Thanks.
Details
Files
feetaway1-000006.safetensors
Mirrors
feetaway1_Numeratic.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1-000006.safetensors
feetaway1_Numeratic.safetensors
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.