How to make videos longer than 15 seconds: stitching multiple Seedance clips

Each Seedance clip generated through RelayDance can be up to 15 seconds long. To make a longer video, split your script into segments of 15 seconds or less and submit each one to POST /v1/video/generations. Reuse the same reference media in every request to keep scenes consistent, download each finished clip, then join the files in order with a video editor or a tool such as ffmpeg.

Why does a long Seedance video need multiple clips?

A single RelayDance video request returns one clip of up to 15 seconds, so longer videos must be built from several requests. Each request includes model, prompt, seconds and a metadata object that holds ratio, resolution, generate_audio, callback_url and content[] reference media. A 60-second video therefore needs 4 segments of 15 seconds, and a 45-second video needs 3. Plan the segment boundaries around natural scene changes, such as a cut, a new camera angle or a change of location, so that the joins look intentional. Use the same ratio and resolution in every request. This lets the clips concatenate cleanly without rescaling or letterboxing. You can check model options and current rates on the RelayDance models page.

What are the steps to generate and stitch the clips?

The workflow is to configure the client, generate each segment, collect the results and concatenate the files in order.

  1. Configure your client. According to the relaydance.com/docs, “Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK”. Authenticate with Authorization: Bearer YOUR_API_KEY, using a key created at relaydance.com/console.
  2. Submit one POST /v1/video/generations request per segment, with seconds set to 15 or less and the same ratio and resolution each time.
  3. Poll GET /v1/video/generations/{task_id} until the status is succeeded or failed. You can also set metadata.callback_url so the final state is POSTed to your server.
  4. Download the video from the url in each result.
  5. Concatenate the clips in script order with an editor or ffmpeg.

How do you keep characters and scenes consistent across clips?

You keep clips consistent by passing the same reference media to every segment request through metadata.content[]. You can attach up to 9 reference images, 3 reference videos and 3 audio tracks per request. Cite each image in the prompt as @image1 to @imageN. A common pattern is to keep a character or product image as @image1 in every segment. You then add a still taken from the closing moment of the previous clip, so the next segment starts from a matching frame. You can also supply the previous clip as one of the 3 reference videos. Reuse the same wording for style, lighting and camera in every prompt. For sound, either set generate_audio per clip or disable it and add one continuous soundtrack after stitching. A single soundtrack avoids audio breaks at the joins.

How much does a stitched 60-second video cost?

A stitched video costs roughly the per-second rate multiplied by its total length, because billing is pay-as-you-go per generated video. The figures below use the published rates for 60 seconds of output:

Retrying a broken segment does not add cost for the failed attempt. According to the relaydance.com docs, “Failed or errored requests are never billed”. Regenerating a clip that succeeded but that you want to replace is billed as a new generation. You can pay with USDT or a Stripe card, and live rates are listed at relaydance.com/models.

FAQ

The answers below cover common questions about stitching Seedance clips on RelayDance.

Can I request a single clip longer than 15 seconds?
No. Each request produces a clip of up to 15 seconds, so longer videos are made by stitching several clips together after download.

Do I need a BytePlus enterprise account to use Seedance?
No. RelayDance provides OpenAI-compatible access to Seedance with no BytePlus enterprise account or KYC required. You create an API key at relaydance.com/console.

Am I charged if one segment fails?
No. Failed or errored requests are not billed, so you can resubmit the failed segment and keep the clips that already succeeded.

According to the official relaydance.com docs: "Failed or errored requests are never billed"
According to the official relaydance.com/docs docs: "Set base_url to https://relaydance.com/v1 and keep your OpenAI SDK"

Key facts and figures

ItemValueSource
Seedance 2.0 720p priceabout $0.190 / secondrelaydance.com/models
Seedance 2.0 1080p priceabout $0.470 / secondrelaydance.com/models
Seedance 2.5 1080p priceabout $0.711 / second, added 2026-10-10, model doubao-seedance-2-5-1080prelaydance.com/models
Seedance Fast priceabout $0.152 / secondrelaydance.com/models
Seedance native 4K priceabout $4.90 per 5-second cliprelaydance.com/models
gpt-image-2 image billingimage output is free, only input is billed; image-to-image from about CNY 0.035; 4K and 1K output cost the samerelaydance.com/models
API protocolOpenAI-compatible; set base_url to https://relaydance.com/v1relaydance.com/docs
Failed requestsfailed or errored requests are never billedrelaydance.com/docs

Data verified 2026-06-29; live prices are on the official /models page.


RelayDance home · Models and pricing · Docs · All guides · Privacy Policy · User Agreement · Telegram community · RelayRouter (LLM API) · Live playground