Wan AI 视频生成器

输入一段文字或上传一张照片,生成带原生音效的短视频 —— 由 Wan 2.5 驱动。浏览器里直接跑,无需安装、无需显卡,注册即送起步额度。

每个新账号都有起步额度 —— 无需绑卡

视频生成器
0 / 2000
消耗 25 积分剩余 0 积分
视频预览

没有生成视频

Made on this site with Wan 2.5 — 5s, 480p, prompt: "a small wooden boat drifting on calm water at sunrise"

Wan 是什么?

Wan 是阿里开源的视频生成模型系列。权重公开,任何人都能下载运行——但跑它需要 GPU 算力,这就是各家托管服务收费的原因。

Open-weight, not a black box

The weights are public. You can run Wan yourself in ComfyUI if you own the GPU — this site just saves you the hardware and the setup.

Native audio, already in sync

Footsteps land when feet land. Speech matches mouth movement. Syncing audio afterwards is the slowest part of short video, and Wan removes it.

Text-to-video and image-to-video

Build a scene from a sentence, or keep your own photo and add motion to it. Image-to-video gives you far more control over the result.

Honest about cost

Wan 2.5 costs about $0.05 per second at 480p and $0.15 at 1080p in real GPU time. We tell you where the free tier ends instead of surprising you.

为什么选这里

把成本讲清楚,把能做什么和不能做什么讲清楚。

Running Wan locally means ComfyUI, model weights and a GPU with serious VRAM. Here it is a browser tab. Generation takes 1–3 minutes.

怎么用

三步出片:写提示词或传图 → 选模型 → 生成。

1

Start from text or a photo

Upload an image to animate it, or skip straight to a prompt if the scene does not exist yet.

2

Describe the motion, not the picture

"Camera slowly pushes in, her hair moves in the wind" beats "a beautiful woman". The model can already see the image — tell it what changes.

3

Pick length and resolution

5 or 10 seconds, 480p or 1080p. Iterate at 480p, then re-run the prompt you like at 1080p. Cheapest way to work.

4

Generate and download

Roughly 1–3 minutes. Audio comes back with the clip, already timed.

能力

文生视频、图生视频,多档分辨率与多个模型可选。

Native audio generation

Sound is produced with the video in a single pass and arrives already synced.

480p and 1080p

Draft at 480p, deliver at 1080p. Roughly a 3× cost difference between them.

5 or 10 second clips

The native length of short-form video. Join several clips for anything longer.

16:9, 9:16 and 1:1

Landscape for YouTube, vertical for Shorts and Reels, square for feeds.

Image-to-video that keeps your subject

Your photo stays recognisable — the model adds motion rather than reinventing the scene.

Open weights

Run it yourself in ComfyUI whenever you want. No lock-in, unlike closed models.

常见问题

What Wan can do, what it costs, and what "free" actually means.








现在就试一次

注册即送起步额度,无需绑卡。