当下最强视频生成模型Sora、可灵AI、Dream Machine对比

OpenAI Sora、快手 可灵AI、Luma Dream-Machine 应该代表了当下最强的视频生成模型。昨天快手的申请终于通过了，用原先Sora公布的提示词做了下对比，感觉真的都还蛮强的。当然只从视频对比Sora无疑是最强的，但是这还一直是一个卖家秀的期货产品，咱们普通人现在也玩不上。另外两者在写实方面的视频表现都还不错，可灵在理解物理世界的表现更强些（汉堡被咬掉），Dream Machine风格相对多样些（会出CG动画的风格），当然这只是从这几个视频表现得出的直观感觉，可能比较片面。

可灵AI可以在快影APP里面申请。Dream Machine 可以在官网直接体验（https://lumalabs.ai/）现在免费版每月可生成30个，30美元套餐一共180个。不得不说视频生成还是很贵的，可灵这两天测试的也不多还没摸到上限。

附提示词：

A stylish woman walks down a Tokyo street filled with warm glowing neon and animated city signage. She wears a black leather jacket, a long red dress, and black boots, and carries a black purse. She wears sunglasses and red lipstick. She walks confidently and casually. The street is damp and reflective, creating a mirror effect of the colorful lights. Many pedestrians walk about.

一位时尚女性走在东京的街道上，街道上到处都是暖色调的霓虹灯和动画城市标志。她身穿黑色皮夹克、红色长裙和黑色靴子，手拿黑色皮包。她戴着太阳镜，涂着红色唇膏。她走起路来自信而随意。街道潮湿而反光，与五颜六色的灯光形成镜面效果。许多行人走来走去。

an older man with gray hair and glasses devours a delicious cheese burger. the bun is speckled with sesame seeds, fresh lettuce, a slice of cheese, and a golden brown beef patty. his eyes are closed in enjoyment as he takes a bite. he wears a red jacket and appears to be sitting inside a fast food restaurant.

一个头发花白、戴眼镜的老人狼吞虎咽地吃着美味的奶酪汉堡。面包上点缀着芝麻、新鲜生菜、一片奶酪和一块金黄色的牛肉饼。他咬了一口，高兴得闭上了眼睛。他穿着一件红色夹克，似乎坐在一家快餐店里。

an alien blending in naturally with new york city, paranoia thriller style, 35mm film

一个外星人自然地融入纽约市，多疑惊悚风格，35mm胶片拍摄。

Cinematic trailer for a group of adventurous puppies exploring ruins in the sky.

一群冒险小狗探索空中废墟的电影预告片

an elephant made of leaves running in the jungle.

树叶做成的大象在丛林中奔跑

A wide shot video of a rally car driving through a redwood forest and showing up behind the trees, spraying dirt into the air as it drifts around a sharp turn, cinematic film shot in 35mm, depth of field.

一段广角镜头视频，拉力赛车驶过红杉林，在树后出现，在急转弯时向空中喷洒泥土，电影胶片采用 35 毫米拍摄，景深