<h2 id="usage-of-v1.0-and-v1.1-189tkivla">Usage of v1.0 and v1.1</h2><p>Trigger words: <code>She is lifting her tops, revealing her breasts and nipples</code> <code>Titty drop, flashing her breasts</code></p><p>Works with <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_720_cfgdistill_fp8_e4m3fn.safetensors">fp8,</a> <a target="_blank" rel="ugc" href="https://huggingface.co/city96/HunyuanVideo-gguf/blob/main/hunyuan-video-t2v-720p-Q5_K_M.gguf">gguf variants</a>, and FastVideo</p><ul><li><p>LoRA strength: 0.5 - 0.8</p></li><li><p>Guidance scale: 9.0</p></li><li><p>Prompt: see the examples</p></li></ul><h2 id="training-details-echbrmxo1">Training details</h2><p>See the "Training Data" file</p><p></p><p></p><h2 id="usage-of-v0-(not-recommended-for-new-users)-t2mumcgwx">Usage of v0 (not recommended for new users)</h2><p>Trigger word: <code>t1ttydr0p</code></p><p>Works with <a target="_blank" rel="ugc" href="https://huggingface.co/Kijai/HunyuanVideo_comfy/blob/main/hunyuan_video_720_cfgdistill_fp8_e4m3fn.safetensors">fp8,</a> <a target="_blank" rel="ugc" href="https://huggingface.co/city96/HunyuanVideo-gguf/blob/main/hunyuan-video-t2v-720p-Q5_K_M.gguf">gguf variants</a>, and FastVideo</p><p><strong>This is still experimental and it seems not activated simply by the trigger word. You need to experiment with seeds and prompts to get the right result.</strong> Here is an example settings that I found it relatively working (also see the example videos and workflows):</p><ul><li><p>num_frames: 41-61</p></li><li><p>seed: 665062624858284 (use this as a starting point)</p></li><li><p>Prompt:</p></li></ul><pre><code>t1ttydr0p, A professional video of a skinny Korean k-pop idol girl dancing on the concert stage. She lifts up her crop top shirt. After a second, she reveals her small breasts and nipples. She wears a crop top and a pleated mini skirt. In the background, there are other girls in the same group fully dressed. navel. Detailed smiling face with gorgeous makeup, high resolution.</code></pre><p>Sometimes, you need to reverse the video. I am not sure how to fix the training process to solve this issue...</p><h2 id="about-the-example-workflows-of-v0-khg6e8a4g">About the example workflows of v0</h2><ul><li><p>HunyuanVideoWrapper (faster, recommended): Optimized for RTX 4090 or above (24GB+ VRAM) with sageattention and triton</p></li><li><p>Comfy: Uses a gguf and the official nodes (optionally you can use sageattention for faster generation)</p></li></ul>

Files with SHA256: c9712691eac408e9452d72c4c75e3f10e5562a248e27bb820440f37c3cccad8d