Start with an image and direct its motion, or switch to text to video when the whole scene should begin from a written idea. The examples below show both approaches with focused five-second briefs.
Seedance 2.0 can begin with either a still image or a written scene. The three examples show object motion, a folding camp stove reveal, and a trail-shoe product launch. Compare what each example starts with, how the camera moves, and where the action ends before choosing an approach for your own shot.
For the folding camp stove, the opening image establishes the orange panels, steel body, mountain setting, and cold light. The motion brief adds a sequence of viewpoints without asking for a new product design. Use this distinction when animating your own object: the image supplies appearance, while the words describe how that appearance should unfold through time. Lead with the main movement. A five-second shot benefits from one coherent action instead of several scene changes.
The alpine-stove example moves from a restrained wide push to a close detail glide and then a low three-quarter hold. Each beat reveals a different part of the same object without asking the stove or setting to transform. Use a short sequence only when every beat has a distinct purpose. For a simpler first attempt, choose one camera move and one clear ending instead of compressing several unrelated views into the clip.
The stove already has a readable silhouette, orange panels, steel parts, and a frost-covered setting. Camera travel and natural highlights create the change while the product remains the visual anchor. When product form matters, name the parts, materials, colors, and background contours that should remain recognizable. This keeps the motion brief focused on viewpoint and light instead of redesign.
The last beat settles into a stable hero view rather than ending during the detail move. Describe where the camera should finish and leave enough time for the product to read at that angle. A deliberate hold makes a short result easier to use in an edit, presentation, or product study. Check that the silhouette, scale, and surface details still read in the closing view.
Choose image to video when a still already defines the composition, subject, or look. Choose text to video when the scene can be invented from words. Both options shown here suit a short, focused shot rather than a long continuous story.
Turn an approved portrait, fashion frame, or character still into a short atmospheric beat. Direct breath, gaze, fabric, hair, particles, or light while protecting identity and composition. Keep the performance small enough to read clearly in five seconds rather than combining dialogue, walking, and a location change.
Test a camera reveal, moving highlight, liquid motion, or environmental change around an existing object concept. Use the result to explore motion and direction. Check logos, labels, measurements, and regulated product claims frame by frame before using the clip publicly.
Add wind, fog, water, traffic, crowds, or gentle camera travel to an environment that is already art-directed. A coherent shot can help establish tone for a storyboard or social concept, while final narrative continuity still requires human selection, editing, and often more than one generation.
The three examples use distinct subjects, environments, palettes, and motion. The two first-frame studies show how a designed composition can stay recognizable through different physical actions, while the text-only example shows how the model establishes a complete scene from a written brief. Use them to decide how much visual control you need before choosing image to video or text to video.
Three examples show both ways to begin a short video.
Choose the workflow first; each option opens with the input it needs.
All three examples use the same focused output settings.
The three films use different subjects, starting inputs, and shot structures.
These limits help you choose the right starting point and judge the result without expecting every possible Seedance setting.
The settings shown here use Seedance 2.0 for five seconds, 720p, 16:9, one output, and no audio. Image to video starts from one source image, while text to video starts from the written brief.
The object-motion study, folding camp stove reveal, and trail-shoe launch are real Seedance outputs. They show three successful directions, but a different subject or prompt may behave differently. Use each example as a visual reference, then compare your own result with the motion you requested.
You can select an image and prepare the motion brief before signing in. The draft stays in this browser and is not uploaded. When online generation becomes available, sign-in will be required before files are sent or credits can be used.
The final credit amount depends on the workflow and settings you choose. It will be shown before generation begins. If generation is unavailable, the page will stop before files are uploaded or credits are used.
Choose Seedance 2.0 in the workspace, select image to video or text to video, and build the brief around the input and settings shown for that workflow.
Select a JPEG, PNG, or WebP image with a clear subject and the intended composition. Choosing a file keeps it in this browser while you prepare the draft.
Write the subject motion, camera behavior, pace, environment, and stability constraints. The current settings stay beside the prompt so you can review the creative direction and output choices together.
Confirm the workflow, input, duration, resolution, aspect ratio, and final credit amount before generation begins. Signing in later will not erase the draft you prepared.
Check subject consistency, motion, camera behavior, text, and the final frame before using the result. If generation fails, the page will show whether you can try again and what happened to the credits.
Plans for image and video
Annual plans are billed upfront. The same plan credits refresh every 30 days and expire at the end of that window; unused subscription credits do not roll over.
Explore the workspace and prepare a first request before choosing a paid plan.
For occasional image and video experiments with a predictable monthly ceiling.
Subscriptions renew automatically at the selected monthly or yearly rate until canceled. Cancel future renewals in your account.
For creators who iterate every week and need room for several directions.
Subscriptions renew automatically at the selected monthly or yearly rate until canceled. Cancel future renewals in your account.
For sustained, high-volume creation and access to the highest available output quality.
Subscriptions renew automatically at the selected monthly or yearly rate until canceled. Cancel future renewals in your account.
Prices are in USD. Subscriptions renew automatically; credit packs are one-time purchases.
Compare plansThese answers cover the image-to-video and text-to-video options shown here, including inputs, output settings, prompts, credits, and current availability.
Seedance 2.0 is a video model used here for image to video with one source frame and text to video with no source frame. Both options shown here create one five-second, 720p, 16:9 video with audio off.
Choose a source image in the generator above, then write what should move over time. Include the primary subject action, camera behavior, pace, environmental movement, and any essential stability constraint. Keep the request focused on one coherent short shot.
Choose 4–15 seconds, 480p/720p/1080p/4K, six frame shapes, and audio on or off. Image to video needs one source image and also offers automatic framing; text to video needs none. Each request creates one video.
Treat the image as the visual starting point and use the prompt for change. A reliable order is subject motion, camera motion, pace, environmental movement, then what must remain stable. Prefer one coherent five-second shot. Specific instructions such as a slow push-in, intermittent wind, or a locked horizon are easier to evaluate than a stack of broad style adjectives.
The final credit amount depends on the workflow and settings you choose. It will be shown before generation begins, and any changed amount will require your confirmation.
If generation cannot start, the page will stop before credits are used. If a started generation later fails, the result screen will show whether it is safe to try again and whether credits are being returned.
Use the Seedance workspace above for an image-led shot, or open text to video and write the scene, motion, camera, and ending from scratch.
