Fakes - Ai Image to Video Disccusion | Page 2 | Social Media Girls

Fakes Ai Image to Video Disccusion

  • Going forward please follow these rules when posting:

    -The content MUST be from TikTok, do not post slips from other platforms.

    -You MUST include profile names / link to the profile, or make every attempt to include it.

    -Include timestamps if you post a long video or a full live video.

    -Make sure ALL girls posted are 18+

  • Please make sure all of the videos and screenshots you're posting are only of people over the age of 18.

    This also means no children in the background.

    No more warnings.

Welcome to the Social Media Girls Forum!
Feel free to sign up and join the discussion on all your favourite Social Media Girls. YouTube and Twitch oops moments, Onlyfans leaks, Celebrity sex tapes and More! It's free and takes 10 seconds!
Sign up
Sep 9, 2025
1
1
Website
Visit site
Try generating on mylovely.ai, here are some examples:
 
  • Like
Reactions: TheBeauCanadian
Comment
Hi there, I wanted to check in to see if anyone have any good, step by step tutorial to do the undress video?

Just to be clear I am not talking about those Img2Vid where you take 1 frame and generate an image to do something. But Vid2Vid where you take one video and change the clothes in the video. Similar to kpopdreamai and kfap.ai on insta




If I understand correctly, you need Comfyui, vace, wan2.1

 
Comment
Does anyone know what is used to make these? Not my videos. used for an example.

 
  • Like
Reactions: mrsnrub23
3 Comments
jinX_4001
jinX_4001 commented
Most likely wan2.2 with some loras higher models produces better fakes tho you need a really good GPU or it could be wan2.5 which was still not available for everyone and it's likely motion muse uses wan2.5
 
D
Duurtyhippy commented
Looks exactly like results I get from a *** bot. You have to pay, but it’s instant, amazing quality and there’s multiple options. All you do is upload a photo, select from blowjob, cumshot, double cumshot, doggy etc. it takes like 2 minutes and you get a 6-15 second video.
 
Last edited:
Z
zulu12 commented
got a link to that bot you mentioned? Found anything better?
 
Grok easy tool
 
3 Comments
W
williwilli commented
Grok got a filter last week. No more NSFW ...
 
Sowmu
Sowmu commented
Not even on supergrok?
 
W
williwilli commented


Nope. I just subscribed few days ago, just to realise that the promised land was closed right before
 

This is by far the best one that I have used....
 
4 Comments
Y
yazir09 commented
yea but what do you do when you run out of free credits
 
R
robstu commented
He's just trying to farm credits from referral..
 
B
bbypocah_ commented


This is actually insulting
 
H
Hamburglar34 commented
wait a day and they refill
 
*** is genuinely the most consistent in terms of making the face stay the same and having a good looking result for Image to video.(if your picture works well enough). It comes preloaded with nsfw format that are consistent if you can't figure out the custom prompt gen. You get 5 credits daily. Custom prompts are 15 credits and the preformatted ones are 10-30.
example stuff i made
 
1 Comment
S
shikler820 commented
un-lu-cid.ai
 

Blinkedtyy

New member
Oct 6, 2025
5
2
Website
Visit site
Does anyone use local generation on their own PC? If so what is recommended to start with. I’ve seen WAN mentioned etc but is there something that’s a complete package or do you need to build it. Thanks
 
2 Comments
H
Hamburglar34 commented
jinX_4001
jinX_4001 commented
Confyui is generally the go to since it can use almost every ai models wan or qwen for image edit also you can just download some workflow for specific ai models that tutorial videos provide tho for me since my PC is not really that powerful I did some tweaking myself and able to do some decent generations if you're planning to try it yourself I would recommend use the comfyui with sage attention for faster generations and lightx2v loras and it also take a lot of space like 200gb for the models and the comfyui itself depending on what you would use