V2.1 Tentacled Wan2.2 - I2V - 14B (Concept)
Optional trigger word: tentaclesex
Describe the visible tentacle. If a random tentacle appears, try reducing the strength of the High LoRA.
I am using both the T2V and I2V versions with the same strength.
The ComfyUI workflow is attached to the sample clip.
or Workflow directly (TrainingData attachement):
https://civarchive.com/api/download/models/2543800?type=Training%20Data
V2.1 Tentacled Wan2.2 - T2V - 14B (Concept)
Optional trigger word: tentaclesex
Example Prompts:
Tentaclesex, woman sitting, tentacle enters her mouth, tentacle blowjob.Tentaclesex, woman laying with spread legs, tentacle enters the screen, tentacle pussy penetration, tentacle moves into her vagina.Tentaclesex, woman laying with spread legs, tentacle pussy penetration, tentacle moving in and out her vagina, tentacle thrusting in and out her vagina.V1 Tentacled Wan2.2 - T2V - 14B (Concept)
The sample videos contain my ComfyUI workflow.
trigger words:
tentacle, tentaclesex, pussy penetration, pussy cumshot, mouth penetration, mouth cumshot, mouthcum, pussycum, cumshot face, cumshot body, fullbody, incomming pussy, incomming mouth
Example Prompts:
tentaclesex, Penetration, Pussy,
A woman with long blond hair, medium skin tone, and medium-sized breasts is lying in a bathtube, legs spread wide. She is naked. Her legs are shiny with a lubricant, and a slimy, dark-colored tentacle is inserted into her vagina. The scene is from a adult film, featuring tentacle erotica.tentaclesex, Penetration, Pussy,
A woman with long black hair, medium skin tone, and medium-sized breasts is lying on a bed, legs spread wide. She wears a black and pink lacy bra and a black skirt. Her legs are shiny with a lubricant, and a slimy, dark-colored tentacle is inserted into her vagina. The background shows a white padded headboard and grey pillows. The scene is from a sci-fi themed adult film, featuring tentacle erotica.Description
FAQ
Comments (36)
Really well made. I'd love to see more ideas from you at this quality level!
is there a way to use it with other character lora while preserving character consistency? I tried block weight but don't know which one has anything to do with face
i think there are no wan2.2 char loras? i have not testet it yet
tawsLoRa 2.1 char loras works on 2.2. btw if I blur faces of some poses like these ones will this lora won't affect other lora's face? check this lora "Character Friendly Body Types - FLUX - Candice-v1-STRONG | Flux LoRA | Civitai".
idk how he prepare dataset and caption it.
@tawsLoRa Training Wan 2.2 character loras is easy and fun https://civitai.com/articles/17740
Nice one
workflow?
on every sample clip - drag and drop into comfyui
I downloaded and tried dragging and dropping your pinned sample clips into comfyui and didn't work. No metadata in the video.
@tawsLoRa it looks like there is no workflow metadata in those samples. Either it was never there or maybe civitai is removing it now
@joannaczarnecka for me its working when i save the video and drag it into comfyui.. but you can use this one from my other lora. https://civitai.com/api/download/models/2137989?type=Training%20Data
@tawsLoRa Much appreciated, thank you
ok this is really well done for some reason, it knows where the orifices' are..... can you use the technique you made for this one and apply to other sex positions, blowjobs etc, what i love the best about it...is that it does not interfere with the faces of characters.
Hi are you using diffusion-pipe for training on wan2.2?
yeah, diffusion-pipe and as models the 28gb
"wan2.2_t2v_high_noise_14B_fp16.safetensors"
"wan2.2_t2v_low_noise_14B_fp16.safetensors"
how do you get the creampie fluid going on, i would ike this for other types of loras
you can try combining it with other loras and describe "thick white fluid cum dripping out of her pussy, slimy" it is trained on a lot of white thick fluid xD
tawsLoRa I see ill try that, thanks
Doesn't seem to work with I2V, anyone manage at all?
If it says T2V then it wont work in I2V
@fox23vang226 wrong. It does.
@m0no1 I'll try and report back
works great with I2V!
I try with both diffusion and GGUF models (NSFW versions), combined with WanImage or PainterI2Vadvanced and basically it works even if i didn't work with all images and it seems it likes a lot pictures with the character seated or half-lenght framed. With other poses i got mixed results (something ok, something horrible). I added the LORA "Fingering" and i was able to get sometimes funny outputs. Having said that, i was unable to get video with the character wrapped by tentacles. And prompting to produce more than 2 tentacles it became very hard to manage via prompt and reach consisten and repeatable results.
Yearning For I2V version!
Is the High lora not the exact same as the Low lora? Same hash, seems to work the same in whatever input too
no they are high and low but i merged both with a little bit of my wan2.1 version, maybe thats why both are the "same"
@tawsLoRa But they have the exact same hash so they are quite literally the same file no? Meaning if you just loaded the "High" lora twice it would be the same as downloading both high and low, saving space
@Kenzato im currently optimizing my dataset for more motion and train a I2V.. maybe i will retrain also the T2V and check the hash to be correct
@tawsLoRa To be clear, you may have made a mistake. If not, then you uploaded the exact same file for high and low Wan2.2-T2V_Tentacled_v1_*_14B.safetensors. They have the same md5 chucksums, meaning they are 100% the same file with different names.
The high and low for Wan2.2-I2V-Tentacled-v2.1-*-tawsLora.safetensors have difference checksums, meaning they are indeed different files.
@tawsLoRa I just realize that I was checking the high lo for v1, not 1.2. For 1.2 T2V high and low have different checksums.
Looks like high and low LoRAs are itentacle. Seems like that might explain the generally subtle movement rather than anything dramatic. LoRA Manager thinks they're both "Low".
no they are high and low but i merged both with a little bit of my wan2.1 version, maybe thats why both are the "same"
@tawsLoRa Mmm, maybe something happened during the merge, or you grabbed the same file twice then? The hash of the high and low versions are the same, which means they are completely byte-for-byte identical.
I calculated the hash for both high and low files on my own machine as well, just to see if the identical hashes listed here on Civit was some sort of weird bug, but both files are identical on my machine too (and the sha256sum from my machine agrees with the CivitAI value.)
@camdynnjovon642 could be, im currently optimizing my dataset for more motion and train a I2V.. maybe i will retrain also the T2V and check the hash to be correct
Details
Files
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Mirrors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentac_v1_LOW_14B.safetensors
T2V_Tentacled_HIGH.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
T2V_Tentacled_LOW.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentac_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
Wan2.2-T2V_Tentacled_v1_HIGH_14B.safetensors
Wan2.2-T2V_Tentacled_v1_LOW_14B.safetensors
T2V_Tentacled_LOW.safetensors
T2V_Tentacled_HIGH.safetensors