Video tool · Live in the studio
AI Text to Video Generator
A sentence in, a moving scene out.
Describe a shot the way you'd describe it to a friend, and get a short video clip back. Camera moves, motion, and light included.
No sign-in needed to look around the studio. Or try the simple version: Wan 2.7 Text to Video.
- Input
- Text
- Output
- Video
- Live models
- 2
- AI lab
- 1
What it does
Video without the shoot
Text to Video runs on Wan 2.2 and Wan 2.7, two of the strongest open video models available right now.
01
Describe the shot
Subject, motion, mood. 'A slow pan across a foggy harbor at sunrise' is a complete instruction.
02
Short clips that loop well
Perfect lengths for social posts, website headers, and mood pieces.
03
Add sound if you want
Wan 2.7 accepts a driving audio track, so your clip can move to your music.
04
Iterate cheaply
Tweak the wording, run it again, compare versions in your gallery.
How it works
How it works
- 01
Write the shot
One or two sentences about what happens.
- 02
Pick Wan 2.2 or 2.7
Fast drafts or highest quality.
- 03
Generate and download
Your clip renders in the studio.
Models
Models
- Wan 2.2 Text to Video
- Wan 2.7 Text to Video
Live in the studio today. Pick any of them from the model menu.
Guide
Prompts that move
Describe the shot like a director, not a novelist. Subject, camera move, and one action beat: "slow push-in on a ceramic mug as steam rises, soft morning window light." Avoid packing five scene changes into one clip.
Short clips are more reliable than epic narratives. Generate, review, then write the next beat as a new job if you need a sequence.
Who text-to-video is for
Use text-to-video when you need a short clip from a written idea: a product motion study, a mood loop, a social bumper, or a concept you cannot shoot yet. Wan 2.2 and Wan 2.7 run here. Wan 2.7 can also take optional driving audio.
If you already have a hero still, image-to-video usually keeps identity better. If music should lead the cut, use audio-to-video.
Practical tips
- One camera move per prompt is enough.
- Name the lighting; it sells realism fast.
- Keep logos and tiny text out of the ask when you can.
- Use audio on Wan 2.7 only when the soundtrack should drive timing.
Know before you spend
Video credits and limits
Video spends more credits than a still. Treat first prompts as tests with clear intent, not long scripts. Morphing faces and melting type still happen; check the full clip before you download.
Browse the app free. Generating needs a subscription so credits can bill against your account.
FAQ
Questions about Text to Video
The things people usually ask before their first generation. Anything else? Ask us.
Length depends on the model and settings in the studio. Plan for short social-length takes, not feature scenes.
Use 2.7 when you want the newer controls, including optional audio. Use 2.2 when you already like its look and want a familiar pass.
