Seedance 2.5
bytedance / seedance-2-5

Seedance 2.5 is ByteDance’s next-generation AI video model, built for up to 30-second video generation, advanced multimodal reference control, precise local editing, and more flexible, production-ready creative workflows.

Model Type:
Pricing: 480P: 17 credits/s ($0.085/s, with video) | 28 credits/s ($0.140/s, no video) 720P: 38 credits/s ($0.190/s, with video) | 63 credits/s ($0.315/s, no video) 1080P: 95 credits/s ($0.475/s, with video) | 158 credits/s ($0.790/s, no video) 🔸Note🔸: Video input has a lower per-second rate because both input and output durations are billed. - With video input: Rate × (Total uploaded video duration + Generated video duration) - Without video input (text/images only): Rate × Generated video duration High-tier top-ups include a +10% bonus, making effective pricing approximately 10% lower than listed. Prices are currently in beta and may be adjusted in the future.
Input
first_frame_url
last_frame_url

Click to upload or drag and drop images.

Images: JPEG/JPG/PNG/WebP/GIF, Max 30MB

The text prompt or description for the video.

0/20480

Click to upload or drag and drop

Supported formats: JPEG, PNG, WEBP, JPG Maximum file size: 30MB; Maximum files: 30

A list of input image URLs.

Click to upload or drag and drop

Supported formats: MP4, QUICKTIME, X-MATROSKA Maximum file size: 200MB; Maximum files: 10

A list of input video URLs. Furthermore, the total length of the three videos must not exceed 30 seconds.

Click to upload or drag and drop

Supported formats: MPEG, WAV, X-WAV, AAC, OGG Maximum file size: 15MB; Maximum files: 10

A list of input audio URLs. Furthermore, the total length of the three audios must not exceed 30 seconds.

Whether to generate AI audio synchronized with the video.

Whether to return the last frame of the video.

The output video resolution.

The aspect ratio of the generated video.

Video duration in seconds.

Video output format.

Enable online search?

Whether to generate a preview draft? Once enabled, it only supports a 480p resolution and allows for quick and cost-effective preview of video effects. Draft task ID is valid for 7 days from draft creation.

A configurable parameter. Defaults to true in the Playground.

There is no guarantee that everything can be filtered out; if you are not satisfied with the results, you will need to make your own arrangements.
Output
output typevideo

README

Complete guide to using Seedance 2.5

Seedance 2.5 API: Build 30s 1080p AI Videos on Kie.ai

Original image
Get Started with Seedance 2.5 API on Kie.ai
Generate Longer 30-Second Videos with Seedance 2.5 API
Native 1080p Output for ByteDance Seedance 2.5 Video Apps
Up to 50 Multimodal References with Seedance2.5
3D White-Model Support for Seedance 2.5 AI Video Generator
Licensed Film IP Creation with Seedance 2.5 AI Model
What Has Improved in Seedance 2.5 Compared with Seedance 2.0 API
Start with Seedance 2.0, Seedance 2.0 Mini, and Seedance 2.5 API on Kie.ai

Seedance 2.5 API provides API access to ByteDance’s upgraded Seedance 2.5 AI video model, designed for longer, higher-quality, and more controllable AI video generation.

Seedance 2.5 on Kie.ai is billed per second of video. Without a reference video, it costs 28 credits ($0.14) per second at 480p, 63 credits ($0.315) at 720p and 158 credits ($0.79) at 1080p. With reference videos, a lower rate of 17, 38 or 95 credits ($0.085, $0.19 or $0.475) per second applies to the total reference-video duration plus the output duration. Reference images and audio add no billable seconds. For worked examples, see our Seedance 2.5 pricing breakdown. Prices may be updated; the pricing card at the top of this page always shows the current rate.

As of its launch announcement on July 31, 2026, ByteDance makes Seedance 2.5 available through its own apps (Jimeng AI and Doubao Pro) and through API access, and the announcement does not include downloadable model weights, so there is no official way to run it locally. On Kie.ai you use it as a hosted API, so you do not need your own GPUs or deployment.

Yes. Seedance 2.5 API is now available on Kie.ai. Developers can access Seedance 2.5 API through Kie.ai and use it for AI video generation workflows, testing, and application development.

Sign in to Kie.ai, create a key on the API Key page (kie.ai/api-key) and send it in the Authorization: Bearer YOUR_API_KEY header. Requests for the model bytedance/seedance-2-5 go to POST https://api.kie.ai/api/v1/jobs/createTask, and the full parameter reference is at docs.kie.ai/market/bytedance/seedance-2-5. Generation is asynchronous: the response returns a taskId; pass a callBackUrl to be notified when the video is ready, or poll GET /api/v1/jobs/recordInfo?taskId=… until the state is success or fail. Generated files are kept for 14 days.

ByteDance officially launched Seedance 2.5 on July 31, 2026, and Seedance 2.5 API is now available for developers through Kie.ai. Developers can access the model and start building AI video workflows with the latest generation capabilities.

Developers can use Seedance 2.5 AI Video Generator for AI video tools, SaaS video platforms, marketing clips, product demos, storytelling workflows, and automated creative production.

Seedance 2.5 API improves video length, native 1080p output, multimodal reference control, and pre-generation planning compared with Seedance 2.0 API and Seedance 2.0 Mini.

Seedance2.5 supports native 30-second AI video generation, helping developers create longer scenes, smoother narratives, and more complete video outputs.

ByteDance Seedance 2.5 supports native 1080p output, giving developers sharper visuals and stronger production quality for AI video applications and creative platforms.

No. Seedance 2.5 outputs 480p, 720p (the default) or 1080p. If you need 4K, Seedance 2.0 on Kie.ai offers 4K output (3840×2160, encoded as H.265) for clips of 4–15 seconds; Seedance 2.5 is the choice when you need clips of up to 30 seconds.

Yes. Kie.ai offers free credits for new users to test supported AI models. Developers can use available credits to access Seedance 2.5 API, evaluate model quality, and test workflow fit before scaling production.

Each video deducts credits from your Kie.ai balance at the per-second rates shown above, and one balance covers every model on Kie.ai. By default, each account can submit up to 20 new generation requests every 10 seconds, which typically allows more than 100 tasks to run at the same time; requests above the limit return code 429 and are not queued. You can also set hourly, daily and total usage caps on each API key.