Smooth Max Wan 视频大模型

Smooth Max Wan 是Smooth Max的文生视频模型版本，基于Wan 2.2 A14B训练而成。我尝试让它尽可能的多才多艺，希望你们喜欢它！展示中的每个视频都使用了我的图库中的一张图片！支持nf/sw，下载包含high和low两个版本以及工作流。

SMOOTHMIX WAN 2.2 T2V v2.0 更新！！！

终于完成了！T_T

SmoothMix WAN 2.2 Txt2Vid v2.0才是该模型最初应该有的样子——现在它终于可以展现真正的实力了！

• 与仅使用 SmoothMix Checkpoints 生成的图像和视频制作的 Loras 合并！
• 图像质量极高，动画流畅！如果您还没有下载，请配合更新后的 Smooth txt2vid 工作流程使用！
• 服装、发型、姿势、体型和肤色种类要丰富得多。
• 您可以使用字幕或提示！两者都可以！请同时使用两者，以确保生成您想要的内容！
• 狐狸精、猫女、恶魔女、鬼女——所有女孩（还有熟女）都在这里。 ;)
• 它响应“ SmoothMixAnime ”和“ SmoothMixRealism ”提示！所有合并到此的Loras都带有SmoothMix动画风格的关键提示，并且在这里也具有相同的效果！详情请查看SmoothMix动画风格页面！
• 它完全未删减，所以应该更适合 NF/SW 内容。不妨试试看。 ;)
• 它无法可靠地生成男性szq！你需要Loras来解决这个问题！SmoothMix的首要任务是服务女性用户。

注意：

• 本次展示的所有视频都是使用我的全新工作流(包含下载) 默认设置制作的。一定要用它！
• 展示视频中没有使用 Loras 制作。请先尝试不使用 Loras 制作视频。
• 使用 Loras 时，首先将其权重设置在 0.3~0.5 之间，然后根据需要增加。

推荐设置

• Steps: 4 or 6
• CFG: 1
• Sampler/Scheduler: Euler a/Normal or UniPC/Simple
• 分辨率：

使用你的系统能够处理的分辨率。首先我推荐以下方法：

• 对于高端规格的电脑：560 x 940 – 940 x 560
• 对于中等规格的电脑：480 x 720 – 720 x 480

测试后，根据需要调整这些分辨率。

作品参照

A photorealistic medium close-up of a young East Asian woman with messy dark hair tied back. She stands on a steel truss bridge during the golden hour sunset. Strong backlighting creates a warm rim light on her hair as it blows in the wind. She wears a brown coat and a dark scarf, her expression is pensive as she looks off-camera. The background shows a city skyline blurred by the setting sun. High definition, cinematic lighting, 8k.
0-1 second: Medium close-up, static camera. A beautiful East Asian woman on a bridge at golden hour, backlit by the sun. Her messy dark hair blows in the wind, she wears a brown coat and scarf, her expression is pensive, looking off-camera.
1-2 second: Medium shot, slow dolly in. The camera pushes slightly closer. She begins to slowly turn her head towards the camera, her windswept hair shifts gently.
2-3 second: Close-up shot, continuing the dolly in. Her face now fills more of the frame as her turn continues, her eyes moving towards the lens. The golden light catches her features.
3-4 second: Close-up shot, slow arc left. The camera begins a subtle orbit around her. She has almost completed her turn, her gaze is about to meet the camera.
4-5 second: Medium close-up, slow dolly out and continuing arc. The camera starts to pull back. A subtle, gentle smile begins to form on her lips as her eyes connect with the lens.
5-6 second: Medium shot, dolly out. The camera pulls back further, revealing more of her coat and the bridge's steel structure. Her soft smile remains, her expression now warm and engaging.
6-7 second: Medium long shot, dolly out with a slight crane up. The camera moves back and slightly upwards, capturing her from the waist up against the bridge and the river below. The cityscape becomes clearer.
7-8 second: Long shot, crane up and away. The camera pulls back into a high-angle long shot, establishing her as a figure on the vast bridge, silhouetted against the beautiful sunset skyline.
Steps: 10, Sampler: Euler, Seed: -1

SmoothMixAnime 真人Cosplay: 《雪山飞狐》中的南兰，南兰青丝绾作高髻，发间缀以珍珠流苏，鬓角斜插一支翡翠步摇，发丝如墨染鎏金光泽。杏眼含黛，眉如远山含黛，眼尾微挑显温婉，眸中流转着琥珀色光晕。瓜子脸轮廓柔和，鼻梁秀气如柳，唇色淡粉似 桃花，颈间玉项链映衬肌肤白皙。身段丰腴却显婀娜，腰肢纤细如柳，肤若凝脂，步履间似踏云霞。淡黄锦袍绣金线云纹，外披素白轻纱披肩，腰系玉带雕龙纹，锦鞋绣金凤头。手持团扇轻摇，扇面绘山水诗画，指尖染墨迹斑驳。气质温婉中透着坚韧，眸光流转 时似有诗词韵律流转，举手投足间似有淡金色光晕萦绕，衣袂翻飞如蝶翼轻颤，步履间似踏着无声韵律。
```yaml
shot:
  composition: "中心构图，中景，突出角色主导地位与视觉张力，通过纵深空间强化情欲氛围"
  lens: "长焦镜头，压缩景深，增强亲密感与身体曲线的立体呈现"
  frame_rate: "60fps，结合慢动作强化动作细节与光影流动"
  camera_motion: "环绕运镜，机位角度为低角度，随角色动作推进形成视觉引导"

subject:
  description: "丰腴曲线与细腻肌肤的完美结合，挑逗姿态中透露掌控欲，眼神魅惑带侵略性，嘴角微扬暗示主动诱惑"
  wardrobe: "丝绸薄纱材质，若隐若现地勾勒身体曲线，腰际系带若即若离，凸显性感妖娆"
  props: "发丝与衣袂的动态互动，配合光影效果增强挑逗感"

scene:
  location: "古色古香的庭院，背景层次分明，前景元素为雕花窗棂与垂花门"
  time_of_day: "黄昏魔法时刻，暖橙色光线穿透窗棂"
  environment: "微风带动衣袂飘动，光线在肌肤上形成动态光斑"

visual_details:
  action: "0-1秒：撩动发丝露出锁骨，顺势转身露出腰际系带；1-2秒：右手抚过胸部，左手轻抚阴户；2-3秒：突然转身，裙摆飞扬，露出大腿曲线；3-4秒：俯身拾起发丝，指尖划过腰际；4-5秒：站起时裙摆如浪，露出臀部线条"
  special_effects: "动作带动光影流动，衣袂与发丝形成视觉漩涡，增强空间穿越感"

cinematography:
  lighting: "侧逆光，突出身体曲线与肌肤质感，形成高对比光晕"
  color_palette: "暖橘与冷蓝对比，强化情欲氛围"
  tone: "禁忌诱惑，融合电影镜头语法强化情色叙事"
```
natural fast motion. real life highly detailed video. the video showcases detailed skin, detailed face, detailed eyes, skin pores.

SmoothMixRealism, A hyperrealistic medium shot of an East Asian woman cosplaying a character with vibrant blue hair in long braids. She wears a leather top, fingerless gloves, and steampunk-style belts, revealing intricate tattoos on her arm and torso. With an intense, direct gaze, she aims a pistol straight at the camera. The setting is a dark, gritty industrial space, dramatically lit with pink and purple hues, creating strong rim lighting and deep shadows. The composition is centered and confrontational, with a shallow depth of field.
0-1 second: Medium shot, slow dolly in. The woman holds the pistol perfectly still, aiming directly forward. Her blue eyes are wide and locked on the camera with a cold, focused intensity. The dark industrial background is softly out of focus.
1-2 second: Medium close-up, the dolly in continues with a slight arc to the left. The camera moves closer, framing her from the chest up. Her grip on the pistol tightens, and the corner of her mouth twitches slightly, hinting at a smirk.
2-3 second: Close-up, the arc continues as the camera pushes in. The frame is now filled with her face and the gun. The smirk grows into a full, manic grin, revealing her teeth. Her eyes gleam with chaotic energy under the pink lighting.
3-4 second: Extreme close-up, snap zoom out. The camera rapidly pulls back from her eye to a point-of-view shot looking down the barrel of the gun, her grinning face now blurred in the background. The focus is razor-sharp on the pistol's muzzle.
4-5 second: Medium shot, static with camera shake. A brilliant muzzle flash erupts from the barrel. The gun kicks back with a powerful recoil, and the entire frame shakes violently for a moment from the percussive blast before settling.
Steps: 10, Sampler: Euler, Seed: -1

A woman in a classroom proudly holds up her drawing of herself. She turns the paper show its other side. The text "Smooth Mix - Wan Checkpoint" is written on it, surrounded by cute drawings. The bright classroom lighting
Steps: 6, CFG scale: 1, Sampler: Euler a

A dark elf assassin waits in a shadowy tavern corner. The camera slowly tilts up, emphasizing her confident smirk.  She adjusts her hair. A single dim light source creates a tense and mysterious mood.
Negative prompt: 色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走, censored, mosaic censoring, bar censor, pixelated, glowing, bloom, blurry, day, out of focus, low detail, bad anatomy, ugly, overexposed, underexposed, distorted face, extra limbs, cartoonish, 3d render artifacts, duplicate people, unnatural lighting, bad composition, missing shadows, low resolution, poorly textured, glitch, noise, grain, static, motionless, still frame, overall grayish, worst quality, low quality, JPEG compression artifacts, subtitles, stylized, artwork, painting, illustration, cluttered background, many people in background, three legs, walking backward, zoom out, zoom in, mouth speaking, moving mouth, talking, speaking, mute speaking, unnatural skin tone, discolored eyelid, red eyelids, red upper eyelids, no red eyeshadow, closed eyes, no wide-open innocent eyes, poorly drawn hands, extra fingers, fused fingers, poorly drawn face, deformed, disfigured, malformed limbs, thighs, fog, mist, voluminous eyelashes, blush,
Steps: 6, CFG scale: 1, Sampler: Euler a

An elegant elven woman plays the piano softly in a candle-lit hall. The camera slowly circles around her, capturing the gentle motion of her hands across the keys. The candlelight flickers, reflecting on her smooth white gown and polished piano surface. Her hair moves slightly with each breath, and the soft glow of the candles enhances the serene, intimate mood. The camera pans upward, showing the tall gothic arches and stained glass faintly illuminated in the background. The focus returns to her calm, confident gaze as she turns slightly toward the viewer, her expression poised and tranquil
Negative prompt: 色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走, censored, mosaic censoring, bar censor, pixelated, glowing, bloom, blurry, day, out of focus, low detail, bad anatomy, ugly, overexposed, underexposed, distorted face, extra limbs, cartoonish, 3d render artifacts, duplicate people, unnatural lighting, bad composition, missing shadows, low resolution, poorly textured, glitch, noise, grain, static, motionless, still frame, overall grayish, worst quality, low quality, JPEG compression artifacts, subtitles, stylized, artwork, painting, illustration, cluttered background, many people in background, three legs, walking backward, zoom out, zoom in, mouth speaking, moving mouth, talking, speaking, mute speaking, unnatural skin tone, discolored eyelid, red eyelids, red upper eyelids, no red eyeshadow, closed eyes, no wide-open innocent eyes, poorly drawn hands, extra fingers, fused fingers, poorly drawn face, deformed, disfigured, malformed limbs, thighs, fog, mist, voluminous eyelashes, blush,
Steps: 6, CFG scale: 1, Sampler: Euler a