Write a prompt or start from a frame, pick Seedance 2 Mini (default), Veo 3.1, MiniMax H3, Wan 3.0 or another model, set the length and resolution, and the AI video generator renders the clip in your browser.
This session
My creationsYour results appear here. Every result is also saved in My creations; if a run fails, the credits go back to your balance.
Seedance 2.5 covers the widest range, from 4 to 30 seconds at up to 1080p. Pick MiniMax H3 for a sharp short shot at 2K, Seedance 2 Mini for quick drafts, and Wan 3.0 when a document or web page should guide the clip.
| Model | Inputs | Resolution | Length | Best for |
|---|---|---|---|---|
| Seedance 2.5Default | Prompt, first and last frames, image, video and audio references | 480p, 720p, 1080p | 4 to 30 seconds | Longer scenes and clips with several references |
| Seedance 2 Mini | Prompt, frames or media references | 480p, 720p | 4 to 15 seconds | Quick drafts before a final render |
| MiniMax H3 | Prompt, first or last frame, media references | 768P, 2K | 4 to 15 seconds | The cleanest short single shot |
| Wan 3.0 Video and Prime | Prompt, frames, media references, a document or a public web page | 480P, 720P, 1080P | 2 to 30 seconds | Clips guided by references or a page |
| Grok Imagine Video 1.5 Preview | Prompt or up to seven images | 480p, 720p | 1 to 15 seconds | Very short loops |
| Veo 3.1 | Prompt, start image and end image | 1080p | Up to 8 seconds | Short cinematic clips in 16:9 or 9:16 |
The credits a run needs are shown before you generate. They change with the model, the resolution and the length you set, and video references add their own seconds on some models.
Write the prompt in five parts: subject, action, camera, light and sound. One action and one camera move per clip keeps the motion steady.
Subject + action + camera + light + sound
A red fox trots through fresh snow at the edge of a pine forest at sunrise. Slow tracking shot at the fox's height, following from the side. Low golden light, breath visible in the cold air. Soft wind and crunching snow, no music. A matte black wireless earbud case opens slowly on a wet slate surface and one earbud rises a few centimetres. Camera holds a close-up, then pushes in slightly. Cool rim light, small water droplets catch the light. Quiet click and a soft low hum. Most clips fall into four jobs. Each card opens the tool or effect page that already has the right setup.
Product clipsStart from a product photo and add a slow orbit or push-in for a store page or ad.Animate a product photo → 
Vertical social clips9:16 at 720p and 5 to 10 seconds fits TikTok, Reels and Shorts.Try Check One Check Two AI → 
Story momentsTurn one photo of a person into a short scene with a set mood and music.Open Day of the Dead AI Video → 
Effect presetsDuets, dances and themed clips with the prompt and settings already filled in.Browse AI effects → Seedance 2 Mini, which has the lowest per-second rate in the model menu. Switch to Seedance 2.5 for 1080p or clips up to 30 seconds, MiniMax H3 for 2K short shots, Wan 3.0 for reference-guided clips, or Veo 3.1 and Grok Imagine Video 1.5 Preview.
Up to 30 seconds with Seedance 2.5 or Wan 3.0, up to 15 seconds with MiniMax H3, Seedance 2 Mini and Grok Imagine Video 1.5 Preview, and up to 8 seconds with Veo 3.1.
Describe the sound in the prompt; models that generate audio will follow it. Seedance 2.5 and MiniMax H3 also accept audio references. You can also add music when you edit or post the clip.
480p, 720p or 1080p with Seedance 2.5 and Wan 3.0, 768P or 2K with MiniMax H3, 480p or 720p with Seedance 2 Mini and Grok Imagine Video 1.5 Preview.
You own what you generate and can use it for personal or commercial projects under our Terms. Make sure you have the rights to any photo, video or audio you upload as a reference.
Usually a few minutes; longer clips, 1080p and 2K take longer. You can leave the page and find the clip in your video history.
Text to video invents the look from your words. Image to video starts from your photo and keeps its subject, framing and style, adding only the motion you describe.
Write a prompt and set the length.
Generate a video