什么是文字生成视频 AI?
文字生成视频 AI 是一种根据文字描述创建视频内容的生成式 AI 工作流。你用提示词描述主体、动作、环境、风格、运镜与光线等,AI 视频模型据此生成连续帧。在 AIImages25 上可选择支持的视频模型,并在生成前配置比例、时长、分辨率与音频等。
文生视频
Members only: video resolutions — upgrade to unlock
消耗: 150 ⚡
示例视频预览
提示词: 暮色中发光萤火虫穿梭的魔法森林
按需求选择可用视频模型与设置。
用文字提示词与支持的模型生成视频。
用文字引导主体、动作、环境、运镜、光线与风格。
文字生成视频 AI 是一种根据文字描述创建视频内容的生成式 AI 工作流。你用提示词描述主体、动作、环境、风格、运镜与光线等,AI 视频模型据此生成连续帧。在 AIImages25 上可选择支持的视频模型,并在生成前配置比例、时长、分辨率与音频等。
从文字提示词开始,模型解释描述并随时间生成画面,尽量呈现主体、运动、环境、风格与运镜。
分析主体、动作、环境、风格等指令。
根据文字描述创建视觉内容。
按动作与运镜在帧之间形成运动。
按所选模型支持的设置合成视频。
不同模型在运动、时长、分辨率、比例、音频等方面不同,请按目标与可用设置选择。
AI 视频模型
ByteDance's latest video model
AI 视频模型
ByteDance's video model
AI 视频模型
Faster & cheaper Seedance 1 Pro
AI 视频模型
Google Veo 3 cinematic video with native audio
AI 视频模型
Faster Veo 3.1 option for supported video generation workflows with audio.
AI 视频模型
Veo 3.1 model option for supported cinematic video generation workflows.
AI 视频模型
Highest-credit Veo 3.1 tier for max quality runs
AI 视频模型
Kling cinematic text/image to video
积分消耗因模型与生成设置而异,请在当前模型界面查看最新要求。
用文字描述主体、环境、动作、运镜与氛围,生成短片。
为短视频场景生成竖版或横版概念。
描述产品、场景与镜头,探索产品视频方向。
为活动与推广生成视觉草稿。
描述角色与运动,探索风格化影像。
把文字想法变成动态草稿,用于分镜与提案。
A lone traveler walks across a vast desert at sunset, wind moving through their clothing as sand drifts across the ground. Wide cinematic landscape, slow tracking camera, warm golden-hour lighting, realistic shadows, atmospheric depth, calm and reflective mood.
A premium perfume bottle stands on a dark reflective surface while soft mist moves around it. The camera slowly pushes toward the bottle, subtle reflections move across the glass, cinematic studio lighting, luxury commercial photography, clean composition.
A fashionable woman walks through a neon-lit city street at night, reflections glowing on wet pavement, natural walking motion, vertical composition, smooth tracking camera, cinematic lighting, energetic urban atmosphere.
A giant floating castle drifts above a mountain valley surrounded by clouds, waterfalls falling into the mist below, birds flying past the camera, slow aerial camera movement, dramatic sunrise lighting, epic fantasy atmosphere.
Keep the camera relatively fixed while movement happens inside the scene.
Describe horizontal camera movement across a scene.
Describe vertical camera movement.
Move visually closer to or farther from the subject.
Follow a moving subject through the scene.
Describe elevated or sweeping movement through a landscape or environment.
选择支持文生视频的模型。
描述主体、动作、环境、风格、运镜与光线。
按用途选择横版或竖版等可用比例。
配置模型支持的时长与分辨率。
在模型支持时启用或调整音频。
开始生成并由模型创建片段。
查看运动与构图,必要时修改提示词或设置。
完成后下载生成结果。
A magical forest with glowing fireflies dancing through ancient trees at twilight, soft volumetric light beams, gentle camera push-in, cinematic atmosphere.
A premium perfume bottle stands on a dark reflective surface while soft mist moves around it. The camera slowly pushes toward the bottle, cinematic studio lighting.
A fashionable woman walks through a neon-lit city street at night, reflections glowing on wet pavement, vertical composition, smooth tracking camera.
A giant floating castle drifts above a mountain valley surrounded by clouds, birds flying past the camera, slow aerial camera movement, dramatic sunrise lighting.
Ocean waves roll onto a rocky shoreline at sunrise, mist rising from the water, slow pan across the horizon, warm natural lighting, calm coastal atmosphere.
A young explorer turns toward the camera in a windy mountain pass, cloak moving with the wind, medium tracking shot, soft daylight, adventurous mood.
Identify the main person, object, animal, product, or environment.
Explain what happens during the clip.
Describe moving water, clouds, wind, smoke, traffic, crowds, particles, or other scene dynamics when relevant.
Add tracking, panning, zooming, aerial movement, or a static shot when the camera direction matters.
Describe time of day, lighting direction, atmosphere, and emotional tone.
For short generated clips, a focused scene with a clear primary action may be easier for a model to interpret than many unrelated events.
在同一平台选择支持的视频模型。
把场景描述变成生成视频。
按模型配置比例、时长、分辨率与音频。
在 AIImages25 内切换图像与视频工具。
从模型目录浏览图片与视频模型。
描述场景、选择 AI 视频模型、配置可用设置,在 AIImages25 上将文字提示词生成视频。