My first Lora Model here for some Veggie fun playing around with cucumbers^^ By using and mixing the tags you might get different poses - in a next version i'll try to focus more on those tags and train few steps shorter.Notes: a) Use a lora weight 0.8 or 0.9 b) Put trigger word "ccmbr" in front of prompts (don't force use "cucumber" in addition, otherwise maybe too many will appear) c) Recommended to use at least min. 3 tags in combination but don't use all tags together d) Generally use minimum CFG scale 7 or try higher !! e.g. by simply using these thre positive prompts "&lt;lora:ccmbr_v1.0:0.8&gt; ccmbr, kneeling, pussy-insertion" and CFG=9 i did get working results with 3 different models in a quick test. It was trained with URPM but it should work with diff. realistic models - but also some anime models worked surprisingly well. Have a look at the last example image grid - there you can see that it works best with weight between 0.7 and 0.9Have fun and let's see your results...

Files with SHA256: 41ddec3b73a6cbe2c9a3ff2c79f6791d0aec723e961587460ac446904b718571