Tentacled
trigger words:
tentacle, tentaclesex, pussy penetration, pussy cumshot, mouth penetration, mouth cumshot, mouthcum, pussycum, cumshot face, cumshot body, fullbody, incomming pussy, incomming mouth
try strength between 0.8 - 1.4
play around with cfg between 5-10 and flowshift 9-12
please share your creations :)
*Workflow attached as TrainingData
Example Prompts:
tentacle, tentacle pussy penetration, hardcore penetration alien tentacle, giant thick tentacle, young naked woman laying with spread legs, direct front view on her vagina, in the bed. tentaclesex.tentacle, tentacle mouth cumshot, blowjob, black slimy alien tentacles, giant thick tentacle, young naked woman laying with spread legs, blindfolded eyes, leather handcuffs, direct front view on her vagina, in the dune desert. tentaclesex.tentacle, tentacle pussy penetration, mouth cumshot, blowjob, black slimy alien tentacles, giant thick tentacle, young naked woman laying with spread legs, blindfolded eyes, direct front view on her vagina, in the dune desert. tentaclesex.tentacle, tentacle pussy cumshot, black slimy alien tentacles, giant thick tentacle, young naked woman laying with spread legs, direct front view on her vagina, skinny thin woman with small breasts, tentacle pit. tentaclesex.Description
Tentacled
trigger words:
tentacle, tentaclesex, pussy penetration, pussy cumshot, mouth penetration, mouth cumshot, mouthcum, pussycum, cumshot face, cumshot body, fullbody, incomming pussy, incomming mouth
- try strength between 0.8 - 1.4
- play around with cfg between 5-10 and flowshift 9-12
FAQ
Comments (8)
the future is now!
idk about the tentacles but your workflow is outstanding. thanks!
We are finally here!
I'm getting this error in comfyui terminal prompt (windows 10) when using your workflow
importlib.metadata.PackageNotFoundError: No package metadata was found for bitsandbytes
what do I have to configure?
switch the textencoder quantization from bnb-nf4 to fp8 or disabled
or optional you can install the bitsandbytes package for 4bit quantization:
copy your comfyui path and open cmd
F:
cd F:\ComfyUI\10_ComfyUI_windows_portable
python_embeded\python.exe -s -m pip install "bitsandbytes"
@tawsLoRa Thanks! I ended pip install, but after that I got this
!!! Exception during processing !!! Can't import SageAttention: No module named 'sageattention'
Traceback (most recent call last):
File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-HunyuanVideoWrapper\nodes.py", line 295, in loadmodel
from sageattention import sageattn_varlen
ModuleNotFoundError: No module named 'sageattention'
I tried by changing attention to Comfy,
and I got this next:
!!! Exception during processing !!! torch._scaled_mm is only supported on CUDA devices with compute capability >= 9.0 or 8.9, or ROCm MI300+
Traceback (most recent call last):
File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 327, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
I'm running a 3090 in windows 10 if that helps.
@nexgenjosi sageattention is a pain to install on windows but its the fastest mode. You can set the model loader to sdpa, this should work but is slower.
you need to use the hunyuan_720_cfgdistill_bf16.safetensors or hunyuan_720_cfgdistill_fp8_e4m3fn.safetensors and set the quantization to "disabled" or "fp8_e4m3fn" i think the "fp8_e4m3fn_fast" only works for 40xx cards.