TheFluxTrain
video Premium

Seedance 2.5 Text

Seedance 2.5 is ByteDance’s longer-form video model on TheFluxTrain. Prompt for cinematic motion, pick duration up to 30 seconds, and generate with native audio.

Quick answer: Seedance 2.5 generates 4–30 seconds of video with native audio from text, a start image, or reference images on TheFluxTrain.

How to use

  1. Open the Seedance 2.5 Text page on TheFluxTrain.
  2. Write a clear prompt describing the result you want.
  3. Pick an aspect ratio that matches your crop (for example 16:9 or 1:1).
  4. Choose an output resolution.
  5. Set the clip duration.
  6. Toggle generated audio on or off if the model supports it.
  7. Click Generate. Sign in if you are not already logged in — credits are used only when generation starts.
  8. Preview the result on this page, then reuse it in other TheFluxTrain studios if needed.

Capabilities

  • Text-to-video and image-to-video with native audio
  • Reference-to-video with up to 30 images, 10 videos, and 10 audios
  • 4–30 second clips at 480p, 720p, or 1080p
  • Wide aspect ratios including 21:9, 16:9, 9:16, and auto

You can draft prompts on this page before signing in. Uploaded media and generation requests are sent only when you start a run, and credits are charged for successful generation jobs.

Frequently asked questions

ByteDance Seedance 2.5 — 4–30s text-to-video + audio
You can open the page and draft a prompt without signing in. TheFluxTrain asks you to sign in when you generate, and uses credits for the run.
Yes. Exploring the page is free. Starting a generation spends credits based on the model and settings you choose. The page shows an estimate before you run.
After you try it here, the same model is available in TheFluxTrain at /video/from-text, and in Flow Studio where that node family is supported.
Yes. Download or reuse the generated clip in TheFluxTrain video editor, Flow Studio, or export it for your own timeline.
Seedance 2.5 Text is marked Premium on TheFluxTrain. Premium models typically cost more credits per run than standard models.
1178 Credits