Wan AI Video Generator
Wan 3.0 in your browser: clips with sound, as sharp as 1080p.

Where it works
Made with Facecam

From text, with café sounds
“Close-up of a barista pouring latte art in a cozy café, the milk swirls into a heart, steam rises, soft morning light, gentle café sounds.”


Before From a picture, with sound
“The chef tosses the vegetables in the pan, flames leap up, he laughs and looks at the camera.”

How to use Wan 3.0
Keep Wan 3.0 picked
Wan 3.0 is already picked on this page; the Model chip switches to any other.
Write the prompt or add an image
Describe the scene, or click + to add a start image and say what happens next.
Make it and download
Pick the length and quality, click Generate, and download the MP4 when it plays.
What is a Wan AI video?
A Wan AI video is a clip made by Wan, Alibaba's AI video model, from a prompt or a picture. Wan 3.0 adds matching sound, goes up to 1080p and handles real, anime and 3D styles.
Wan 3.0 features
Sound included
Ambience, effects and music with the picture.
720p or 1080p
1080p is the sharpest clip in the tool.
5 or 10 seconds
Wide, tall or square from text.
Real, anime or 3D
Say the style in the prompt.

Wan 3.0 on Facecam
| Wan 3.0 | |
|---|---|
| Made by | Alibaba |
| From | Text, or a start image |
| Lengths | 5 s, 10 s |
| Quality | 720p, 1080p |
| Shapes from text | 16:9, 9:16, 1:1 |
| Sound | Yes, made with the picture |
| Under the hood | Wan 3.0 |
| Our clips took | About 1 to 2.5 minutes (5 s, 720p) |

More about Wan 3.0
Wan AI Video Generator is an online tool that runs Wan 3.0, Alibaba's video model, to make a 5 or 10 second clip with sound from a text prompt or a start image, at 720p or 1080p.
Wan 3.0: what you need and limits
| How it works | Online. Works on any laptop. |
|---|---|
| You get | Video download |
| You need | Any up-to-date browser |
| Phones and tablets | Supported |
| Make it from | A text prompt (text to video), or a start image plus an optional prompt (image to video). |
| Models | Veo 3.1, Seedance 2.0, Kling 3.0, Gemini Omni Flash, Grok Imagine 1.5, Veo 3.1 Fast, Seedance 2.0 Fast, Wan 3.0, Veo 3.1 Lite and Hailuo 2.3. All but Hailuo make sound; on Kling and Veo 3.1 it's your choice. |
| Length | 4 to 10 seconds, depending on the model. |
| Quality | 720p (Hailuo 768p); 1080p on every model but Seedance 2.0 Fast and Hailuo. |
| Frames | A start image, and on Veo, Kling, Seedance 2.0 and Gemini Omni an end image too. |
| Shape | Wide 16:9, tall 9:16 or square 1:1 from text; a start image keeps its own shape. |
| Time | About 1 to 3 minutes, depending on the model and length. |
| Price | Comes with every paid plan, by the clip; each model, length and quality has its own price, shown before you make it (e.g. 31 Kling 3.0 clips of 5 seconds a month on Premium). A clip that fails doesn't count. |
| Output | An MP4. Free results have a small Facecam mark in the corner; plans remove it. |
| Kept for | 7 days in your Library, then deleted. Start images are deleted within an hour. |
Getting good Wan clips
Wan rewrites short prompts into fuller ones before it makes the clip, so a single clear sentence is enough: “a barista pours latte art in a cozy café, close-up, soft morning light”. Name the style if it isn't a real-looking scene (anime, 3D animation, watercolor) and say what you want to hear. Choose 1080p when the clip will be watched on a big screen; 720p is quicker and uses less of your plan.
Your images and videos
A start image is stored privately while the video is made and deleted when it's done, always within an hour. The video stays in your Library for 7 days, then it's deleted.

Wan 3.0 FAQ
How do I use Wan AI?
Wan 3.0 is already picked on this page: write a prompt or add a picture, choose 5 or 10 seconds and 720p or 1080p, and click Generate.
How much does Wan 3.0 cost?
It comes with every paid plan, by the clip: 26 Wan 3.0 clips of 5 seconds a month on Premium, if you use nothing else. Longer clips and higher quality take more; the tool shows what each one takes before you make it, and a clip that fails doesn't count.
Does Wan make sound?
Yes. Wan 3.0 clips come with sound made for the scene.
Is this Wan 2.1 or Wan 3.0?
Wan 3.0, the newest Wan model, with sound and up to 1080p.
How long does Wan take to make a clip?
Between one and two and a half minutes for 5 seconds at 720p in our tests. Ten seconds or 1080p adds time; the finished clip is saved to your Library either way.
Can Wan turn a photo into a video?
Yes. Add a start image and Wan adapts the clip to the picture's own shape, then moves it the way you describe. The sneaker turntable on the image to video page was made this way, sound included.
Should I pick 720p or 1080p on Wan?
Pick 1080p for anything watched full screen on a TV or a laptop, or when you'll crop in later. For phone feeds and quick tests, 720p looks fine and leaves more of your plan.
Can Wan make anime videos?
Yes. Start the prompt with “Anime style:” and Wan holds that look from frame to frame. The anime rooftop scene with wind and piano on the AI Video Generator page is a Wan 3.0 clip.
Related
- Seedance 2.0 Video GeneratorSeedance 2.0 in your browser: cinematic shots that come with their own sound, up to 1080p.
- Hailuo AI Video GeneratorHailuo 2.3 in your browser: 6 or 10 second clips that stretch your plan furthest.
- Text to VideoWrite one sentence, get a video: any scene, any shot, in a minute or two.

Try Wan 3.0 now
- Cancel any time
- Runs in your browser
- No gaming PC needed





