<p><span style="color:rgb(34, 139, 230)">If you want to give runpod credits to help with training, feel free to send a code in dm.</span></p><p>A version of the furry nsfw lora for the 14b wan model. <strong>Also works with humans.</strong></p><p><span style="color:rgb(250, 82, 82)">This is an img2vid lora, it's meant for the wan 14b img2vid models (both 480p and 720p should work). Attempting to generate txt2vid can yield unexpected results.</span></p><p><span style="color:rgb(250, 82, 82)">Avoid using teacache, if you do use it, keep the threshold low. Teacache causes more artifacts with this lora, sometimes making it do strange things.</span></p><h3 id="(up-to)-5x-speedup-with-causvid-rqqx524xu"><span style="color:rgb(34, 139, 230)">(Up to) 5x speedup with causvid</span></h3><p>I recommend using <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan21_T2V_14B_lightx2v_cfg_step_distill_lora_rank32.safetensors">lightx2v causvid</a>, combined with the <a target="_blank" rel="ugc" href="https://huggingface.co/alibaba-pai/Wan2.1-Fun-Reward-LoRAs/blob/main/Wan2.1-Fun-14B-InP-MPS.safetensors">mps reward lora</a> for more movement. When combined, complicated actions continue to have lots of movement. Even on 6 steps with <strong>euler/euler a</strong> + <strong>beta</strong> it yields great results with lots of motion and great physics.</p><p>Previous causvid lora info below (Personally I prefer v1 at 50% over v2)</p><p>Using the <a target="_blank" rel="ugc" href="https://civarchive.com/models/1585622?modelVersionId=1794316">causvid lora</a> you can get a speedup of around 5 times (assuming you used 20 steps beforehand) It even works with the wan gguf models. Just make sure to follow these steps:</p><ul><li><p>Load the causvid lora at 0.5 strength (as well as this lora at 1 strength)</p></li><li><p>Sample at 4 (minimum) to 8 (6+ seems to be the sweet spot) steps with a low cfg (&lt;=3, I usually use 2), <strong>use beta scheduler for best results</strong>. If you see ghosting, ensure your steps and sampler are set correctly.</p></li></ul><p>On my 3060, this lets me generate a 60 frame video in under 4 minutes. The quality is usually higher than teacache, and it's much faster. Do not combine with teacache.</p><p></p><h2 id="purpose-qvaj5lcz0">Purpose</h2><p>This lora can keep characters consistent, and can handle many positions from pov or similar perspectives. It was trained on many positions, follow the prompting guide. Avoid using with t2v loras, your characters might warp and transform.</p><p>This lora is capable of generating (without the need for other loras): cowgirl(+reverse), missionary, doggystyle, blowjob(+deepthroat)</p><p>It is also capable of handjobs, titfuck. v1 might need assistance, v1.1 seems to be good on that.</p><p>It is effectively a lora for NSFW motions in i2v without changing character consistency. And with a better understanding of furry characters.</p><h2 id="v1.1:-8o6ehib5h">V1.1:</h2><p>Continued training with a new dataset, entirely new captions, more perspectives. It usually yields more motion, and it's easier to tag. It can do many positions, perspectives and motions without the need for a second lora.</p><p>Prompting v1.1 is like prompting a t2i model, aside from motions, you can prompt for "moving up and down" or similar. Although v1.1 will usually have motion anyways.</p><h2 id="v1:-snd6hkanl">V1:</h2><h3 id="note:-if-you're-not-getting-enough-motionnot-the-right-speed-2mazmgxfe">Note: if you're not getting enough motion/not the right speed</h3><p>If you don't prompt for motion, you won't get any.</p><p>"deep thrusts, fast thrusts", "medium sucking, slow sucking", etc. will adjust the depth and pace.</p><p>"The woman moves up and down as she rides the man", "the woman uses her breasts to stroke the man's cock." Should be self-explanatory. You can even prompt for pulling out, varying degrees of success.</p><p>Prompting should be similar to prompting the <a target="_blank" rel="ugc" href="https://civarchive.com/models/1429979/furry-nsfw-wan-21-13b-img2vid-wan21fun-inp">1.3b version</a>.</p><p>Trained on 400 res, frame buckets of [1, 8, 16, 24, 32, 40, 60, 80], context as "multiple_overlapping".</p><h3 id="compared-to-1.3b-version-y989vdnll">Compared to 1.3b version</h3><p>This model performs much better at oral, has less stretching artifacts. But might be a little harder to prompt right for the motion, at least on really short videos. Make sure you include a speed and depth in your prompts in the case it doesn't animate enough, this could help.</p><h3 id="trained-on-img2vid-not-recommended-for-txt2vid-zs0ou8ber">Trained on img2vid, not recommended for txt2vid</h3><p>I cannot give any promises about quality when used on txt2vid. Especially furry content, I have not tested it and cannot guarantee quality. It might be able to generate some human content, maybe a little furry content. But I would recommend using a different lora instead for those situations.</p>

Furry nsfw wan 2.1 14b img2vid (and nsfw motion in general)

<p>Resumed from epoch 17 lora, with a new dataset, 22 new epochs, so technically epoch 39, although the epochs with the new dataset were slightly shorter, it has more variety.</p>

Files with SHA256: 65e7478a5a871eaddcf12abc3af526695135edab471cb982d0fb4253bcabf24a