<h2 id="i-didn't-know-you-could-milk-a-bull"><strong>I didn't know you could milk a bull 🤨</strong></h2><p>Don't worry babe, just trust me 😇</p><hr /><h2 id="training"><strong>Training</strong></h2><ul><li><p>360 steps on high noise</p></li><li><p>3 edited videos (40 frames each video)</p></li><li><p>ZCD (These acronyms were created solely to confuse you)</p></li><li><p>Lots of confidence</p></li><li><p>$4 (✋🏻🤠🤚🏻)</p></li></ul><hr /><h2 id="usage-(low-noise)"><strong>Usage (Low noise)</strong></h2><p>This LoRA was trained only on High Noise, so you should load a Low Noise LoRA that knows what a penis is (In this case I recommend DR34ML4Y v2)</p><hr /><h2 id="uncensored-prompts-and-examples"><strong>Uncensored prompts and examples</strong></h2><p>You know that Civitai likes to remove some LoRAs, right? For that reason, I won't post the "Cut to action" prompt. To see all my LoRAs, <a target="_blank" rel="ugc" href="https://huggingface.co/Aisha-AI-Official/models">visit my repository on HuggingFace.</a></p><p>Or <a target="_blank" rel="ugc" href="https://huggingface.co/Aisha-AI-Official/wan2.2-milking-table">visit this LoRA on HuggingFace 🤗</a></p><hr /><h2 id="i2v:"><strong>I2V:</strong></h2><p>Ongoing:</p><pre><code>She moves her hand up and down repeatedly, stroking the hard penis.</code></pre><p>Prompt 1 (cut to action):</p><pre><code>Heeeey cowboy, not here... ✋🏻🤠</code></pre><p>Prompt 2 (cut to action):</p><pre><code>Heeeey cowboy, not here... ✋🏻🤠</code></pre><p>Prompt 3 (cut to action):</p><pre><code>Heeeey cowboy, not here... ✋🏻🤠</code></pre><p>High Noise LoRA Scale: 1</p><p>Low Noise LoRA Scale: 1</p><p>Shift: 4</p><hr /><h2 id="t2v:"><strong>T2V:</strong></h2><p>Theoretically it works, but I haven't tested it. If you want to test it, keep the same structure as I2V, and start with a low scale on High Noise (0.5).</p><hr /><h2 id="about-zcd-(zero-caption-dropout)"><strong>About ZCD (Zero Caption Dropout)</strong></h2><p>Caption dropout is a technique that allows some steps to be learned without the prompt. When LoRA is used separately, this isn't a big problem, but it can cause LoRA to kick in even if you don't ask for it.</p><p>This is bad when you try to incorporate LoRA into a checkpoint (you've probably already asked for a simple woman in an NSFW Flux checkpoint and ended up getting a woman with nipples showing through her shirt. That's what happens when you do caption dropout training 🤷🏻‍♀️)</p><h3 id="negative-effects:">Negative effects:</h3><ol><li><p>It will probably only work at the same camera angle</p></li><li><p>High chance of not responding to different prompts</p></li><li><p>High chance of forcing the original characters of the training video (reduced if you train only the high noise)</p></li></ol><hr /><h2 id="download"><strong>Download</strong></h2><p><a target="_blank" rel="ugc" href="https://huggingface.co/Aisha-AI-Official/wan2.2-milking-table/resolve/main/milking_table_high_noise.safetensors"><strong><u>Download High Noise LoRA (On HuggingFace 🤗)</u></strong></a></p><p><a target="_blank" rel="ugc" href="https://huggingface.co/datasets/Aisha-AI-Official/sdxl-loras/resolve/main/DR34ML4Y_I2V_14B_LOW_V2.safetensors"><strong><u>Download Low Noise LoRA (DR34ML4Y_I2V_14B_LOW_V2)</u></strong></a></p><hr /><h2 id="want-to-request-a-lora!"><strong>Want to request a LoRA?!</strong></h2><p><strong>(Or help making a new one)</strong></p><p>Just visit <a target="_blank" rel="ugc" href="https://huggingface.co/Aisha-AI-Official/LoRA-requests"><strong><u>LoRA Requests</u></strong></a> and see if what you want has already been requested, or start a new request for it.</p><hr /><h2 id="help-me-creating-more"><strong>Help me creating more</strong></h2><p>If you want to help me continue making LoRAs, or if you want me to make a LoRA for you, buy 5000 PlayCoins at <a target="_blank" rel="ugc" href="https://beta.aisha-ai.com/"><strong><u>Aisha-AI</u></strong></a> and transfer them to my account (account number 2).</p><p>This helps Aisha-AI to stay alive and produce new LoRAs for you all 💜</p>

Files with SHA256: f15379155d43b7732f9c8432490446efc4683f70cadac5bd9ee509c9326773f8