Seedance 2.5 Flash Text to Video API

seedance-2.5-flash/text-to-video

Seedance 2.5 Flash Text to Video turns text prompts into 4–30 second videos with synchronized audio, auto or six fixed aspect ratios, and 480p or 720p output. It follows the action order, camera moves, and sound cues you write, keeping characters and settings coherent across the clip.

Input

691/20000

Seedance 2.5 Flash Text to Video

Seedance 2.5 Flash is the lighter, lower-cost tier of ByteDance Seed's Seedance 2.5 video model and keeps its multimodal audio-video workflow. Text to Video builds a whole scene from language alone, generating picture and sound in one pass for clips from 4 to 30 seconds. Its 480p and 720p output fits drafts, social video, and high-volume concept work where cost per second matters.

Why choose the Seedance 2.5 Flash Text to Video API?

  • About a third less per second720p runs at 41 credits per second and 480p at 18.5, compared with 63 and 28 on Seedance 2.5, so long clips and many iterations cost less.

  • Up to 30 seconds in one generationChoose any whole-number length from 4 to 30 seconds and let a scene develop through several beats in one continuous clip.

  • Picture and sound togetherTurn on generate_audio to add ambience, effects, and dialogue that match the action, at the same per-second rate as a silent render.

  • Framing for every channelPick auto or one of six fixed ratios from 21:9 to 9:16 to compose for widescreen, vertical, or square delivery.

  • Same request as Seedance 2.5The input fields match Seedance 2.5 Text to Video, so you can prototype on Flash and switch the model ID when a shot needs 1080p.

Parameters

ParameterRequirementDescription
promptRequired

String. Describes the subject, action, camera, lighting, and sound.

Default—
durationRequired

Integer from 4 through 30 seconds, inclusive; the Playground preselects 5.

Default—
resolutionOptional

Output resolution; the Playground always sends it explicitly.

Default720p480p
aspect_ratioOptional

Output framing; the Playground preselects auto.

Default—auto16:99:161:121:94:33:4
generate_audioOptional

Boolean. true generates synchronized audio and false returns a silent video at the same rate; the Playground sends true by default.

Default—truefalse

How to Use

  1. Describe the sceneOpen with the subject, setting, and premise: a cook lifting noodles from a steaming pot at a night stall.

  2. Order the actionWrite beats in sequence with words such as first, then, and finally so the clip has a clear progression.

  3. Direct the cameraAdd shot size, angle, and movement in a separate sentence, for example a steady medium shot that drifts closer.

  4. Add look and soundFinish with lighting, palette, and the sounds you want, such as bubbling broth and distant chatter.

  5. Set the outputChoose duration, resolution, aspect ratio, and audio, then check the price shown on the Run button.

  6. Run and refineReview picture and sound together, then adjust the prompt where timing or framing needs another pass.

Pricing

Price is set by output duration and resolution; audio on or off uses the same rate.

UsageRateDetails
480p18.5 credits/output sec ($0.0925/sec)4 seconds costs 74 credits ($0.370), 10 seconds costs 185 credits ($0.925), and 20 seconds costs 370 credits ($1.85).
720p41 credits/output sec ($0.205/sec)4 seconds costs 164 credits ($0.820), 10 seconds costs 410 credits ($2.05), and 30 seconds costs 1,230 credits ($6.15).

Best Use Cases

  • Draft concept filmsTest a campaign or story idea as moving footage at 480p before committing to a higher-resolution render.

  • Social video at volumeProduce vertical and square variations of one idea for different channels at a lower per-second rate.

  • PrevisualizationTurn a scripted beat into a motion reference for reviewing pacing, staging, and camera direction.

  • Mood and music visualsGenerate atmosphere pieces where the sound design and visuals are planned in the same prompt.

Pro Tips

  • Use this prompt order: subject and setting, ordered action, camera direction, lighting, then sound.
  • Keep subject movement and camera movement in separate sentences so each instruction has one job.
  • Iterate at 480p to settle timing and framing, then render the final pass at 720p.
  • Describe the sounds you want by source, such as rain on a canopy or footsteps on gravel, when audio is on.
  • Build each short clip around one clear visual idea.

Notes

  • Tasks run asynchronously: a successful submission returns a task_id for polling or callback delivery.
  • For 1080p output, send the same input fields to Seedance 2.5 Text to Video.

Related Models

Seedance 2.5 Flash Text to Video API frequently asked questions

What is the Seedance 2.5 Flash Text to Video API?

Seedance 2.5 Flash Text to Video is a ByteDance Seed model for generating video from text prompts. It creates 4–30 second clips at 480p or 720p with optional synchronized audio, auto framing, and six fixed aspect ratios, at about a third less per second than Seedance 2.5. Built on the Seedance 2.5 multimodal audio-video workflow, it follows the action order, camera direction, and sound described in the prompt while keeping characters and settings coherent. You can call it programmatically or try it from the playground above.

How does Seedance 2.5 Flash Text to Video compare with Seedance 2.5?

Both endpoints accept the same input fields and generate 4–30 second clips with optional synchronized audio. Flash outputs 480p or 720p at 18.5 and 41 credits per second, while Seedance 2.5 Text to Video costs 28 and 63 credits per second and adds 1080p. Draft and publish at 720p on Flash, and switch the model ID to Seedance 2.5 when a shot needs 1080p.

Can Seedance 2.5 Flash Text to Video generate a 30-second clip?

Yes. Set duration to any whole number from 4 to 30 seconds. Billing follows the output length, so a 30-second clip at 720p costs 1,230 credits ($6.15). Describe the beats in order so a longer clip develops clearly from start to finish.

How many credits per second does Seedance 2.5 Flash Text to Video use?

480p uses 18.5 credits per output second ($0.0925) and 720p uses 41 ($0.205). A 5-second 720p clip costs 205 credits, and audio on or off uses the same rate. The Run button in the playground shows the exact price before you submit.

Does Seedance 2.5 Flash Text to Video generate synchronized audio?

Yes. Set generate_audio to true and the video comes with sound generated in the same pass, such as ambience, effects, and dialogue. Name the sources you want in the prompt for clearer results. Setting it to false returns a silent video at the same price.

Which aspect ratios does Seedance 2.5 Flash Text to Video support?

Choose auto or one of six fixed ratios: 21:9, 16:9, 4:3, 1:1, 3:4, and 9:16. Use 9:16 for vertical social video, 16:9 or 21:9 for widescreen scenes, and 1:1 for square feeds.

How should I structure a Seedance 2.5 Flash Text to Video prompt?

Write the subject and setting first, then the action in order, then camera direction, lighting, and sound. Keep camera movement in its own sentence and use words such as first, then, and finally for multi-beat clips. Iterating at 480p is a cost-efficient way to settle timing before the 720p pass.

When should I use Seedance 2.5 Flash Image to Video instead of Text to Video?

Choose Seedance 2.5 Flash Image to Video when you already have the opening frame, such as a product shot or character design, and want the video to start from it. Stay with Text to Video when the whole scene can be described in words.