Seedance 2.5 is ByteDance's multimodal video generation model, and doubao-seedance-2-5-260628 brings text creation, image-driven generation, audio and video references, video editing, and extension together in a single workflow. It is suitable for projects requiring more complete action sequences, multiple reference assets, or adaptation of existing videos, and can generate videos up to 30 seconds long at up to 1080p, with or without sound.
Submit requests to the public API at api.acedata.cloud using the documented parameters, then use the results in your application.
Input parameters and result formats vary by service. Use the public API for this model and follow its guide for generation, task retrieval and editing operations.
Clarify capacity, inputs and outputs, and invocation methods before choosing a model.
The above are the applicable specifications for this model on this platform. When invoking it, select the corresponding settings based on generation, editing, or extension mode.
Learn what doubao-seedance-2-5-260628 can bring to your work.
Images, videos, and audio can all participate in creation, or you can provide only audio references. When organizing assets, clearly specify in the text which images define the subject, which videos are used for actions or camera movement, and which audio is used for sound and rhythm, so that multiple references work toward the same goal rather than simply piling up assets.
In addition to generating videos from text or images, 2.5 also provides edit and extend modes, allowing you to request modifications or extensions based on existing videos. It is suitable for separately stating “what to keep” and “what to change,” for example, keeping the subject and camera movement, changing the atmosphere of the sky, or continuing to describe subsequent actions.
The maximum duration of 30 seconds is suitable for arranging an opening, main actions, and a conclusion, reducing the need to split a complete creative idea into too many short segments. When generate_audio is enabled, sound can be generated at the same time; when using audio references, you should still clearly describe the visual actions and sound intent to make it easier to review rhythm and expression together.
Start with specific tasks to find where the model can be effective.
Enter product images, usage steps, and shot descriptions, use reference images to define the subject, and organize continuous sequences around presentation, use, and closing shots. Vertical format is suitable for short video ads, while horizontal format is suitable for demonstration content; after delivering the video, focus on checking whether the product form and actions meet requirements, then create subtitles and brand information.
Enter character images, action videos, or audio references, describe the scene, action sequence, and sound atmosphere for the character, and generate video segments suitable for story previews or content planning. If the creative idea is primarily driven by sound, you can also provide only audio references and text descriptions without first producing image or video assets.
Use an existing video as reference_video, and clearly specify the elements to modify or the actions that need to continue. edit is used to modify existing content, while extend is used to lengthen clips; choose MP4 or MOV as the delivery format, and retrieve results through asynchronous tasks for inclusion in subsequent editing and asset management workflows.
Choose based on task complexity, input materials, and expected results.
Compared with the 4–15 second Seedance 2.0 series, 2.5 extends duration to 4–30 seconds and adds pure audio references as well as explicit editing and extension modes. If the task includes longer action sequences, coordination of multiple assets, or adaptation of older videos, 2.5 is a better fit; existing short-shot workflows can retain the original model for comparison.
2.5 supports up to 1080p, so it should not be understood as exceeding 2.0 in every specification. For 4k output, consider 2.0 Standard; for short-video iteration requiring only 480p or 720p, compare 2.0 Fast or Mini. When choosing, first assess whether the complete action is usable and whether the assets are effective, then determine whether the output resolution meets delivery requirements.
From a small-scale task to formal integration.
Define the objective, required inputs, and output requirements, using real business examples as a starting point.
Open the trial page, confirm the parameters supported by this entry point, then submit a small-scale task to review the results.
Keep the complete model ID, use the request format specified in the documentation, and confirm billing rules on the Pricing page.
Example inputs and review methods to help you design your first trial.
Start with a reference image of a person or product, and describe the subject's action, environment, and camera movement. Example prompt: “Maintain the subject's appearance from the reference image, use one continuous shot, and have the camera slowly move forward; the subject completes one clear action, while the background and lighting remain stable, with no scene changes.” Video duration is set separately in the API.
Check whether the subject remains consistent across frames, whether the action is completed, whether the camera movement is coherent, and whether the ending feels natural. When multiple scenes are needed, first define the goal of each shot, then separately check transitions and pacing; when sound is needed, configure and preview it according to the audio settings supported by the interface.
Before formal use, understand the output quality and capability scope.
Answers to common questions when using doubao-seedance-2-5-260628.
This model supports 480p, 720p, and 1080p, with 1080p as the maximum. If delivery explicitly requires 4k, consider Seedance 2.0 Standard; if the main needs are longer clips, audio-only references, editing, or extension, then 2.5 is more suitable, so there is no need to choose based only on version number.
Yes. 2.5 supports using reference_audio alone, without also providing an image or video. Pair it with text describing the visual subject, action, and scene; using an audio reference in creation does not mean sound output is automatically enabled. When sound is needed, also set generate_audio:true.
Choose edit to modify existing content and extend to continue expanding a clip; both require a reference video and adaptive aspect ratio. The duration for edit must be -1; extend can use 4–30 seconds or -1. Prompts should respectively emphasize the content to change or the action that needs to continue.
first_frame and last_frame are used to specify the starting or ending image, while reference_image provides reference information such as the subject. Use the first frame when you want a video to begin moving from an image; when you want to create a new scene or action based on an image, use a subject reference and explain its purpose in the text.
Submit model and content to /seedance/videos, with model set to doubao-seedance-2-5-260628. Set async:true to first obtain task_id, then query the task status; you can also set callback_url to receive completion notifications, and finally get the finished video from the video link in the result.
Model information · Updated: 2026-10-01. For invocation parameters and billing rules, see the API and pricing sections.
Choose a development approach or application suited to this model and follow its setup guide.
↗Python SDK Integration Guide
SDK installation, authentication and examples↗TypeScript SDK Integration Guide
SDK installation, authentication and examples
↗
↗
↗Pricing
$0.11–$1.44· per second
You pay$13,584for 142,800 credits · $0.0951 / credits46% OFF
You're billed in credits. Pick a top-up tier to see its per-credit rate — the larger the top-up, the cheaper each credit.
This is only a basic call example. See the full docs for more parameters and advanced usage.
View Docs