Preparing page…
▥ Seedance 2.5 is here! ›
PricingDocsUpdatesAccount

bytedance/seedance-2-5

ByteDance's next-generation video model for longer cinematic clips, multimodal references, keyframes, and synchronized audio.

Warm Official Pre-release Priced by route Data privacy Zero training API docs

Input

prompt*string

Describe the subject, motion, camera, lighting, and timing

route_modestring

Choose the provider route used for this request.

Lower-cost official route and the default for new requests and API keys.
generation_typestring

Select the generation workflow. The API mode updates with this choice.

aspect_ratiostring

Aspect ratio of the generated output

Default: "adaptive"
resolutionstring

Higher resolutions may take longer to generate.

Default: "1080p"
durationnumber

Length of the generated video in seconds

generate_audioboolean

Generate synchronized audio with the video.

184 credits · $0.92

Output

Seedance 2.5 sample output

Generated in 38.6 seconds

Examples

View more examples ↗

Pricing

Choose the same three routing tiers used across MLMux. The selected route is included as route_mode in every request.

Learn more about pricing ↗

Readme

Seedance 2.5 is available through MLMux with one request shape and three explicit provider routes.

Longer cinematic shots

Generate clips up to 30 seconds for more complete narrative sequences, product films, and social campaigns.

Multimodal control

Combine text, start and end frames, images, videos, and audio to control identity, motion, timing, and camera language.

Production output

Use asynchronous predictions, webhooks, and explicit routing to move from prompt tests into reliable production workloads.

Model details

Owner
ByteDance
Type
video generation
References
Up to 9
Default route
official-cheap

More video models

Explore all models ↗
bytedanceSeedance 2.0

ByteDance's multimodal video model for text, keyframes, image, video, and audio references with stable subjects and natural motion.

bytedanceSeedance 2.0 Fast

ByteDance's faster Seedance 2.0 model for short-form video generation, prompt iteration, keyframes, and multimodal image references.

bytedanceSeedance 2.0 Mini

ByteDance's lightweight video model for high-throughput, cost-sensitive generation with text, keyframes, references, and synchronized audio.

One API for image and video generation.

ModelsPricingDocumentationConsole