Based on some suggestions in addition to those of my generous Commissioner @shin_gekiganger , I am here to present this pose.
Note : pony version is not recommended
It has some flows but it gives solid results most of the time
----------------------------------------------
it is trained on 2 main views for the hold : Front View, Side view and each one have a trigger word
trigger words :
global trigger: TriChoke
Pose trigger :
1) FrontViewChoke
2) SideViewChoke, closeupSideVew ( this makes more closeups )
Helping words :
legs wrapped around the neck and arm
Armgrabbing
styles : Illustration,faceless ,3DModelstyle
Example (front view)
- masterpiece,best quality,amazing quality,closeup view, TriChoke++, FrontViewChoke++,on the floor ,karate+ gi , 1girl, 1boy, girl ( cute girl, middle hair, red hair, thighs sports bra) legs wrapped around the neck and arm, wrestling, hetero, faceless male+, bold, between thighs,blush, indoor dojo,Armgrabbing
Example 2 (side view )
masterpiece,best quality,amazing quality, prespective long shot, trichoke++, SideViewChoke, closeupSideVew, 1girl, 1boy, black hair, hetero, buruma, blonde hair, short hair, gym uniform,leg wrapped around the neck++,Armgrabbing+
-------------------------------------------------------------------------------------------------------
I specialize in making minor and forgotten characters, and challenging concepts , if you like my work, consider supporting me :
HanyoAI - Professional, Digital Artist | DeviantArt
also, feel free to share your NSFW images on my server
https://discord.gg/P4BNMYhd
Description
FAQ
Comments (7)
These are all very nice and appreciated as they are NOT easy to create, LoRA-wise. Any chance for a keylock/ude-garami/Americana? My favorite submission when I was grappling, also, might be tough to do from both side mount and mount. But worth a try.
this one is little tricky still possible ...thanks for the suggestion
Pony version is planned?
okay why not it is easy
@HanyoAI thank you its what i miss and what controlnet cant do properly
@magmirant0011 ControlNet is accurate but need a huge time and manipulations; anyways i already trained it
@magmirant0011 i think making a good robust pony version out of this might take time .. it works but it is far inferior compared to the other version
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.

