# Seedance 2.0

`bytedance/seedance-2.0` · ByteDance · video

Direct a shot using text, opening and closing images, or visual and audio references. Each input has a specific role, so an image can set the first frame or guide a subject that appears later.

[Model page](https://sociaro.com/models/bytedance/seedance-2.0) · [JSON](https://sociaro.com/models/bytedance/seedance-2.0.json) · [Model pricing](https://sociaro.com/pricing?model=bytedance%2Fseedance-2.0) · [Documentation](https://docs.sociaro.com/models/bytedance-seedance-2-0/) · [Playground](https://my.sociaro.com/playground/?model=bytedance/seedance-2.0) · [Compare Seedance API pricing](https://sociaro.com/pricing/seedance)

Video capabilities — `video-generation`, `video-edit`

## Use cases

- Animating between approved first and last frames
- Building a shot from separate location and character references

## Technical details

- API model ID — bytedance/seedance-2.0
- Provider — ByteDance
- Modality — Video
- Input — Text prompt · Images · Video · Audio
- Output — Video · Image
- Endpoint — POST /v2/jobs
- Workflow — Omni video
- Resolution — 480p · 720p · 1080p · 4k
- Aspect ratios — 21:9 · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 · adaptive
- Duration — 4–15 seconds

## Request parameters

- `resolution` — Choose the video resolution.
  - Allowed values — `480p`, `720p`, `1080p`, `4k`
- `duration` — Choose 4–15 seconds, or use -1 to let the model set the duration.
- `ratio` — Set the frame shape, or choose adaptive to let the model match the inputs.
  - Allowed values — `21:9`, `16:9`, `4:3`, `1:1`, `3:4`, `9:16`, `adaptive`
  - Input images and reference videos must be 300–6,000 pixels on each side and have a width-to-height ratio from 0.4 to 2.5. Each video frame must contain 409,600–8,295,044 pixels.
- `output_format` — This model returns MP4 video.
  - Allowed values — `mp4`
- `generate_audio` (boolean) — Generate sound in sync with the video. Sound is off by default unless you supply an audio reference.
- `bitrate_mode` — Choose the encoding bitrate. Leave this unset to use the model default.
  - Allowed values — `standard`, `high`
- `watermark` (boolean, default `false`) — Whether to put an AI-generated mark in the lower right corner of the clip; off unless true.
- `seed` (integer) — Seed for the generator, a whole number from -1 to 2147483647; the same seed gives similar, not identical, results.
  - -1–2147483647
- `return_last_frame` (boolean, default `false`) — Return the final frame as a separate image alongside the video, at no extra charge.
- `execution_expires_after` (integer) — Set a deadline for the generator in seconds. The default is 48 hours. The job can still time out sooner.
  - 3600–259200
- `safety_identifier` (string) — A stable identifier for the end user. Sociaro sends the provider a pseudonym.
  - 1–64 characters
- `content` (array, required) — Put the prompt in a text part. For media, put a public HTTPS URL or data URI in image_url.url, video_url.url or audio_url.url. Choose keyframes or reference media for each request.
  - Allowed values — `text`, `image_url`, `video_url`, `audio_url`
  - At least 1 item
  - Start with first_frame and optionally add last_frame, or use up to 9 reference_image inputs. Each image must be under 30 MB. Use JPEG, PNG, WebP, BMP, TIFF, GIF, HEIC or HEIF.
  - Use reference_video for up to 3 MP4 or MOV clips. Each clip must be 2–15 seconds, 24–60 fps and at most 200 MB. All input video together can run up to 15 seconds.
  - Use reference_audio for up to 3 WAV or MP3 clips. Each file must be 2–15 seconds and at most 15 MB. Combined audio can run up to 15 seconds. Audio needs a reference image or video.
  - For an edit or extension, refer to source clips as [Video 1], [Video 2] and so on, in input order. To join clips, supply two or three in sequence.

## Pricing

Rates in USD. Catalogue timestamp 2026-11-21T02:23:44Z.

A 5-second 720p clip at 16:9 without input video costs about USD 0.83. Billing is token-based.

- USD 7.7 per 1M tokens · 480p · active
- USD 4.73 per 1M tokens · 480p, with input video · active
- USD 7.7 per 1M tokens · 720p · active
- USD 4.73 per 1M tokens · 720p, with input video · active
- USD 8.47 per 1M tokens · 1080p · active
- USD 5.17 per 1M tokens · 1080p, with input video · active
- USD 4.4 per 1M tokens · 4K · active
- USD 2.64 per 1M tokens · 4K, with input video · active
- USD 0.077339 per second of output video · 480p · active · approx., 16:9
- USD 0.047508 per second of input and output video · 480p, with input video · active · approx., 16:9
- USD 0.16632 per second of output video · 720p · active · approx., 16:9
- USD 0.102168 per second of input and output video · 720p, with input video · active · approx., 16:9
- USD 0.411642 per second of output video · 1080p · active · approx., 16:9
- USD 0.251262 per second of input and output video · 1080p, with input video · active · approx., 16:9
- USD 0.85536 per second of output video · 4K · active · approx., 16:9
- USD 0.513216 per second of input and output video · 4K, with input video · active · approx., 16:9
- USD 0.362267 minimum per job · 480p, with input video · active · for a 4 s output; more for a longer one
- USD 0.780267 minimum per job · 720p, with input video · active · for a 4 s output; more for a longer one
- USD 1.913511 minimum per job · 1080p, with input video · active · for a 4 s output; more for a longer one
- USD 3.901333 minimum per job · 4K, with input video · active · for a 4 s output; more for a longer one

- Metered in tokens, so the charge follows the frames rendered; the per-second figures are an estimate for 16:9.
- A job with input video is metered at the with-video rate, and is charged at least the minimum per job in its own row.
- Leaving out resolution gives 720p.

## Examples

### Above the ridgeline

An aerial camera follows a red paraglider along a high mountain ridge.

[Result (video)](https://previews.sociaro.com/shared/promo/gallery/alibaba-video/paraglider.mp4) · [Poster](https://previews.sociaro.com/shared/promo/gallery/alibaba-video/paraglider.webp)
