Get the latest global workflows on my RunningHub homepage and run them free on RTX4090.<ul><li>Sign Up with my invite link for 1000 bonus points + 100 daily:</li><li><a target="_blank" rel="ugc" href="https://www.runninghub.ai/?inviteCode=rh-v1221">https://www.runninghub.ai/?inviteCode=rh-v1221</a></li></ul><ul><li>Run Directly:</li><li><a target="_blank" rel="ugc" href="https://www.runninghub.ai/post/1956586031082602497/?inviteCode=rh-v1221">https://www.runninghub.ai/post/1956586031082602497/?inviteCode=rh-v1221</a></li><li>Please credit and link to my homepage if reprinting.</li></ul>The model only needs to load high noise lora, recommended weights: 0.8-1, to get better water quality, you need to turn off light acceleration at low noise, increase the number of steps, and improve the CFG value (3.5).Example prompt word:yanmo567, masterpiece, best quality, photorealistic, full shot, a woman in a grey dress and black thighhighs bending over towards a small kitten on a newspaper, then water suddenly rushes up from the bottom and submerges the entire screen.yanmo567, a woman in a green sweater and a cap takes a selfie in an elevator, and as the doors open, a huge splash of water erupts from the bottom and submerges the entire screen.

wan2.2-i2v-high-Screen Flood（屏幕淹水）

淹没-high.safetensors

high

The model only needs to load high noise lora, recommended weights: 0.8-1, to get better water quality, you need to turn off light acceleration at low noise, increase the number of steps, and improve the CFG value (3.5).该模型只需要加载高噪声lora，推荐权重：0.8-1，要获得更好的水的画质，需要在低噪声下关闭light加速，增加步数，提高CFG值（3.5）。Example prompt word:yanmo567, masterpiece, best quality, photorealistic, full shot, a woman in a grey dress and black thighhighs bending over towards a small kitten on a newspaper, then water suddenly rushes up from the bottom and submerges the entire screen.yanmo567, a woman in a green sweater and a cap takes a selfie in an elevator, and as the doors open, a huge splash of water erupts from the bottom and submerges the entire screen.

Files with SHA256: fffc73e51a967a51f40bd4ed8661d1bd42c5e5f32adca4f8fbbf638ce88c0cea