Text to video
Describe the subject, action, setting and camera movement. Keep the prompt focused, then adjust one detail at a time.
SEEDANCE 2.5 GUIDE
Learn what Seedance 2.5 can do, how to shape a prompt and which settings to check. This is an independent workspace, not an official ByteDance or Dreamina website.
MODEL CAPABILITIES
Use a prompt and supported references to describe the subject, movement, camera, lighting and sound. The available controls depend on the selected model.
Describe the subject, action, setting and camera movement. Keep the prompt focused, then adjust one detail at a time.
Add a suitable image when you want to guide the opening composition, subject or product.
Where supported, combine image, video or audio references and state what each one should guide.
Some modes offer audio controls. Check the selected model’s settings to see which options are available.
MODEL COMPARISON
Model versions differ in supported inputs and output limits. Always check the model name and current settings in the generator.
ByteDance Seed’s published launch information describes text, image, video and audio inputs, with clips up to 15 seconds. The settings available in this workspace may differ.
ByteDance Seed says Seedance 2.5 supports clips up to 30 seconds and up to 30 images, 10 video clips and 10 audio clips per request. Check the workspace for the controls available here.
Only use the versions shown in the model picker. For release announcements, check ByteDance Seed’s official model pages.
For official specifications and release updates, use ByteDance Seed’s own pages. This workspace is an independent third-party service.
A PRACTICAL WORKFLOW
A useful result starts with a clear brief, references you have permission to use and a careful review of the output.
Write down the subject, action, setting and camera movement before adding optional visual details.
Use images, footage and audio only when you have the rights and consent needed for your project.
Check the crop, captions, branding and destination platform requirements before you share a generated clip.
PROMPT GUIDE
Name the subject and action first, then add the setting, camera movement, lighting and sound. Keep instructions specific and easy to follow.
Subject + action + setting + camera + lighting + sound
At blue hour, a cyclist rides along a rain-slick city street. The camera tracks beside them as warm shop lights reflect in the pavement; distant traffic and soft rain are audible.
THREE-STEP WORKFLOW
Select text to video or image to video, depending on how you want to start.
Describe the scene in English or another language supported by the model. Review duration, resolution, audio options and the credit estimate.
Check the result, then change one instruction at a time so you can see what improved.
IDEAS TO TRY
FAQ
Start with the subject and action, then describe the setting, camera movement, lighting and sound. Use clear instructions and avoid asking for several conflicting actions at once.
Use text to explore a new scene. Start from an image when you want to guide the composition, subject or product. Check which mode the selected model supports.
ByteDance Seed publishes a maximum duration of 30 seconds for Seedance 2.5. This workspace may offer a smaller set of durations or resolutions; use the controls shown before you generate.
Any trial credits or offers can change. Check the balance in your account and the credit estimate shown before you generate; no permanent free allowance is promised.
No. This is an independent service that provides access to supported models. It is not operated by ByteDance, Dreamina or the Seedance model team.
The releases have different inputs, audio features and output limits. ByteDance Seed describes Seedance 2.0 as supporting clips up to 15 seconds and Seedance 2.5 as supporting clips up to 30 seconds. Check the provider’s documentation for full specifications.
ByteDance Seed published its Seedance 2.0 launch announcement on 12 February 2026. Use the official announcement linked on this page for the original release information.
Choose the version shown in the model picker, select a supported mode and add a prompt or references. Review the current settings and credit estimate before submitting.
Availability can change. Check the model picker for the versions currently enabled in this workspace, and use ByteDance Seed’s official pages for release announcements.
Read the message from the selected provider and make sure your prompt and references follow its safety rules. Remove material you do not have rights to use; changing a prompt does not guarantee acceptance.
Practical guides
Find a clear walkthrough, prompt examples and up-to-date notes on credits before you start.
A five-step workflow covering inputs, prompts, settings, credit estimates and how to refine a result.
Read the guideSeven adaptable examples for portraits, product videos, short-form clips, image-to-video and sound.
Read the guideCheck starter credits, generation costs, watermarks and privacy before you pay.
Read the guideChoose a supported mode, write a prompt and review the available settings and credit estimate before you generate.
Start creating