Video model by ByteDance
Write a scene or add a first frame, and Seedance 2.0 turns it into a 4 to 15 second clip with sound, in any of six shapes.
Examples
Clips Seedance 2.0 made from a written prompt or one picture, played as they came out.
01Shot lists it can follow
Seedance 2.0 reads a prompt the way a crew reads a shot list. Write the shots in order, say where the camera starts and where it goes, and the clip tends to keep that order instead of blending everything into one shot. Short, separate beats such as “wide shot, then close-up on the hands” work better than one long paragraph.
02Steady camera moves
Push-ins, slow pans, orbits around a subject and handheld follow shots are where Seedance 2.0 is comfortable. Name the move and its pace in plain words, like “slow dolly in” or “the camera circles the table once”, and keep one main move per shot. Framing holds well, which matters for product shots and calm establishing scenes.
03Sound made with the picture
Seedance 2.0 renders audio together with the video, so footsteps land on the step, the rain you hear matches the rain you see, and a door sounds when it shuts on screen. Describe the ambience, the effects and the mood of any music in your prompt. If you plan to add your own track later, switch audio off before you create.
04Every shape, even from a photo
Six aspect ratios are available in both modes: 9:16 for TikTok, Reels and Shorts, 16:9 for YouTube and slides, 1:1 and 3:4 for feeds, 4:3 for a classic frame and 21:9 for a wide cinematic strip. Unlike the other two models on this site, Seedance 2.0 lets you pick the ratio even when you start from an image.
05Bringing a still to life
Add one image as the first frame and Seedance 2.0 animates forward from it: a product photo starts to turn, a portrait looks up, a drawn scene begins to move. The first frame fixes the look, the colours and the composition, so your prompt only has to describe what happens next and how the camera behaves.
06Two quality settings
Choose 480p to try an idea or check the timing, then 720p for the version you post. Both settings cover every length from 4 to 15 seconds and every aspect ratio. A 480p run uses fewer credits, so shape the prompt there first and rerun the same settings at 720p once the motion looks right.
Seedance 2.0
Shot-by-shot scenes with sound, in any of six shapes
Describe the subject, the action, the camera and the sound in a few short sentences. For image to video, drop a photo on the first-frame slot; Seedance 2.0 starts from that picture and your prompt says what moves next.
Open the settings and choose one of six aspect ratios, a length from 4 to 15 seconds, and 480p or 720p. Leave audio on to get sound with the clip, or turn it off if you will add music later.
Press Create; the button shows the credits for these settings before anything is spent. Sign in and top up only if you need to, then download the finished MP4. Every clip is also saved in My videos.
Seedance 2.0 is a video generation model made by ByteDance. It turns a written prompt, or a prompt plus one starting image, into a short video clip with sound. On Copy That Video you can run it right on this page, or pick it from the model menu on the text-to-video and image-to-video tools, next to Seedance 2.5 and MiniMax H3.
Each clip runs between 4 and 15 seconds. The length menu offers 4, 5, 6, 8, 10, 12 and 15 seconds. Credits are charged per second, so a shorter clip uses fewer of them. If you need one continuous take longer than 15 seconds, Seedance 2.5 goes up to 30 seconds with the same two quality settings.
Seedance 2.0 renders at 480p or 720p. You can choose 9:16, 16:9, 1:1, 3:4, 4:3 or 21:9, and that choice stays available in image-to-video mode as well. For the cleanest result from a photo, upload one that already has roughly the shape you pick. Use 480p for drafts and 720p for the clip you share.
Yes. Audio is generated together with the picture: ambient sound and effects that match what happens on screen, and music if you ask for it. You can steer it in the prompt, for example “quiet café chatter, a cup set down on a saucer”. The audio switch in the settings bar turns it off when you would rather add your own soundtrack.
Start from text when you want the model to design the whole scene. Start from an image when the look has to be exact: a product, a character you drew, a place you photographed. The image becomes the first frame and the prompt describes the motion and the sound. Both modes offer the same lengths, qualities and aspect ratios.
Each run is priced per second of video, and the total depends on the length and quality you pick. The Create button always shows the credits for your current settings before you pay anything. Credits come in one-time packs and never expire, and if a render fails, the credits go back to your balance automatically.
Your image is stored only so the model can read it, and it is deleted as soon as the render ends, whether the clip succeeds or fails. The finished video is saved to My videos in your account, so you can watch and download it again later. Please upload only images you have the right to use, and no photos of other people without their consent.
All three make clips from text or from a first frame. Seedance 2.0 runs 4 to 15 seconds at 480p or 720p and keeps the aspect ratio choice in image mode. Seedance 2.5 stretches to 30 seconds at the same two qualities but follows the photo’s shape in image mode. MiniMax H3 renders in 2K, runs 4 to 15 seconds and always includes sound.
Same prompt box, different limits. Switch any time from the model button.
