Felo API PlatformFelo API Platform
Back to models

Seedance 2.5 Flash API

ByteDance's Seedance2.5 Flash delivers fast multimodal video generation with native audio, reference controls, editing, and multilingual lip sync.

ByteDance's Seedance2.5 Flash is a multimodal generation model accepting text, images, video, and audio. It supports text-to-video, image-to-video with first-frame control, first-and-last-frame control, multimodal reference-to-video generation, video editing, and video extension. It can generate ambient sound, sound effects, and music together with video in a single pass, while audio-only references can guide music-matched visuals.

The model is designed for high-volume production and rapid iteration, including social media videos, product animations, scene transitions, and international audiovisual content. It supports multilingual audiovisual generation with lip sync in20 languages, optional web search in text-to-video mode, and reference inputs of up to10 images,5 videos, and5 audio clips per request. Video duration ranges from4 to30 seconds, with480p and720p output, multiple aspect ratios, and MP4 or MOV formats.

Seedance2.5 Flash is approximately twice as fast as Seedance2.5 while offering quality close to the flagship model. It does not include draft mode, and its output resolution is limited to480p and720p. Audio generation can be toggled on or off, and the model can return the last frame as an image.

Updated:

MODEL ID

seedance-2.5-flash
Output
video
Compare all Seedance models

Highlights

  • Joint audio-video generation
  • First-and-last-frame control
  • Multilingual lip sync
  • Video editing and extension
  • 480p and720p output
  • Optional web search

Providers

1 providers

Compare API prices across providers. Select a provider to view all charges and conditions.

Provider · Pricing details
Uptime

Benchmarks

  • End-to-end latency P50

    seconds; the best result among providers

    72.59
  • Generation speed

    seconds of video per minute of generation time; the best result across providers

    5.0
  • Arena.ai against Veo

    Ranking verified on September17,2026.

    #6
  • Arena.ai against Kling

    Ranking verified on September17,2026.

    #4
  • Arena.ai against Runway

    Ranking verified on September17,2026.

    #2

Scores as published by the model provider.

Video specifications

Aspect ratios
16:9 / 4:3 / 1:1 / 3:4 / 9:16 / 21:9
Audio · Output
Yes
Resolutions
480p / 720p

API reference

Model-specific protocol details are not published. Check the API reference for integration requirements.

API reference →

Supported parameters

  • prompt
  • duration
  • resolution
  • aspect_ratio
  • generate_audio
  • seed
  • size
  • frame_images
  • input_references
  • user
  • session_id
  • trace

Frequently asked questions

What is Seedance 2.5 Flash?

ByteDance's Seedance2.5 Flash is a multimodal generation model accepting text, images, video, and audio. It supports text-to-video, image-to-video with first-frame control, first-and-last-frame control, multimodal reference-to-video generation, video editing, and video extension. It can generate ambient sound, sound effects, and music together with video in a single pass, while audio-only references can guide music-matched visuals. The model is designed for high-volume production and rapid iteration, including social media videos, product animations, scene transitions, and international audiovisual content. It supports multilingual audiovisual generation with lip sync in20 languages, optional web search in text-to-video mode, and reference inputs of up to10 images,5 videos, and5 audio clips per request. Video duration ranges from4 to30 seconds, with480p and720p output, multiple aspect ratios, and MP4 or MOV formats. Seedance2.5 Flash is approximately twice as fast as Seedance2.5 while offering quality close to the flagship model. It does not include draft mode, and its output resolution is limited to480p and720p. Audio generation can be toggled on or off, and the model can return the last frame as an image.

How is usage billed?

Your bill depends on the provider you use, its prices and your usage. Longer inputs or other conditions may have different prices. Billed based on the actual generation cost. See your bill for the final amount.

Which inputs does Seedance2.5 Flash support?

It supports text, images, video, and audio. A request can include up to10 images,5 videos, and5 audio clips as reference inputs.

What context and output lengths are supported?

The context limit is20,480 characters. Video duration can be set from4 to30 seconds or set to -1 for automatic selection, with a default of5 seconds.

What tasks is this model suited for?

It is suited to high-volume generation, rapid iteration, social media content, product animations, multilingual audiovisual content, video editing, and video extension.

How does it differ from Seedance2.5?

Seedance2.5 Flash is approximately twice as fast as Seedance2.5 and offers quality close to it. It accepts fewer reference inputs and has no draft mode.

What video and audio controls are available?

It supports first-frame and first-and-last-frame control, native audio generation, audio-only references for music-matched visuals, multiple aspect ratios, and MP4 or MOV output.

Related models

Seedance 2.0 Mini

ByteDanceVideoseedance-2.0-mini

Seedance 2.0 Mini is ByteDance's compact, cost-efficient video generation model for high-volume short clips, with text, image, video and audio references.

Concept testingStoryboard draftsBulk variant generationSocial ad testing

Resolution

480p / 720p

Duration

—

Billed based on the actual generation cost. See your bill for the final amount.

Seedance 2.5

ByteDanceVideoseedance-2.5

ByteDance Seed's Seedance 2.5 generates up to 30-second continuous video with synchronized audio from text, images, video and audio references.

Short drama productionAdvertising filmsProduct demos and trailersOnboarding and training videos

Resolution

—

Duration

—

—

Seedance 2.0

ByteDanceVideoseedance-2.0

ByteDance's Seedance 2.0 is a natively multimodal video foundation model that generates synchronized audio and video from text, image, clip, and audio inputs.

Multi-shot narrative film productionMarketing and social ad campaignsE-commerce product showcasesEducational training videos

Resolution

480p / 720p / 1080p / 4K

Duration

—

Billed based on the actual generation cost. See your bill for the final amount.

Seedance 2.0 Fast

ByteDanceVideoseedance-2.0-fast

ByteDance's speed-optimized Seedance 2.0 Fast video model generates 4–15 second clips up to 1080p with synchronized audio from text, image, video, and audio inputs.

Rapid video prototypingSocial media creative variantsAdvertising draftsStoryboard visualization

Resolution

480p / 720p

Duration

—

Billed based on the actual generation cost. See your bill for the final amount.