ByteDance: Seedance 2.5

Seedance 2.5 extends multimodal video creation into longer storytelling, with clips up to thirty seconds. It is useful for scenes that need more room for a narrative beat, alongside reference-driven creation, editing, and continuation.

Compare models

seedance-2.5

AvailableByteDanceVideoAsyncNew

Input / Output: $0.00 / $10.70
Released: Jun 28, 2026
Modalities: Video

About Seedance 2.5

Seedance 2.5 is ByteDance Seed's audio-video generation model aimed at thirty-second storytelling, the successor to Seedance 2.0. A single pass can run up to thirty seconds, and clips can be extended further. ByteDance adds timestamp-level control for editing audio and video, plus green screen, camera perspective and reference-based edits. It keeps the reference, extension and video-to-video operations of 2.0.

Where it works well

Clips up to thirty seconds in one pass, in which ByteDance says several connected shots can build setup, development and resolution.
Timestamp-level control for targeted edits to audio and video content.
Advanced editing such as green screen, camera perspective and reference-based edits, according to the launch post.
Mov as well as mp4 output, which helps when footage goes straight into an editing suite.
Reference-driven creation and continuation inherited from the 2.0 generation.

When to choose another model

Output is offered at 480p and 720p, so 1080p or 4k delivery needs Seedance 2.0.
Long clips give more room for drift, so character and setting details should be repeated in the prompt.
Newer than 2.0, with fewer public examples to learn prompting patterns from.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Image to videoReference to videoStart and end frames to videoText to videoVideo extensionVideo to videoTokenLab endpoint

POST/v1/videos/generations

curl -X POST "https://api.tokenlab.sh/v1/videos/generations" \
  -H "Authorization: Bearer sk-xxx" \
  -H "Content-Type: application/json" \
  -d '{
  "aspect_ratio": "adaptive",
  "duration": -1,
  "operation": "text-to-video",
  "output_audio": true,
  "output_format": "mp4",
  "resolution": "720p",
  "model": "seedance-2.5",
  "prompt": "Cinematic sunrise over a calm lake with gentle camera motion."
}'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

480p / 720p · No video input

per 1M tokens

Official price: Output $10.70
Verified price: —
Discount: —

480p / 720p · Includes video input

per 1M tokens

Official price: Output $6.40
Verified price: —
Discount: —

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
480p / 720p · No video input	Output $10.70	—	—
480p / 720p · Includes video input	Output $6.40	—	—

This table shows per 1M tokens unit prices. Actual video cost also depends on generated token usage, which is affected by resolution, duration, and whether video input is included.

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 12, 2026, 09:33 PM UTC

Metrics appear once privacy and data volume thresholds are met.

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Seedance 2.5 in Console with a prompt ready to edit or send.

Help me create with seedance-2.5 using text-to-video at /v1/videos/generations. Show the result, status, and cost.

Use cases

Best for

Text to Video

One-take short stories
Write a complete beat structure, from setup to payoff, and generate it as a single thirty-second piece with cuts included.
Timed audio-video revision
Fix one moment in a clip by editing the section at a given timestamp rather than rerunning the whole take.
Green screen and relight work
Replace backgrounds or change camera perspective on existing footage in preparation for ad and film work.

Prompt examples

0-10s: a courier races through rain with a parcel. 10-20s: he reaches a locked door and hesitates. 20-30s: it opens to a surprised child.

Keep the dialogue at 12 seconds, but change the music underneath from piano to a low synth pulse.

Remove the background behind the speaker and place her in front of a calm office window.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Seedance 2.5?

ByteDance Seed's next-generation audio-video model after Seedance 2.0. It is built for thirty-second storytelling with precise reference control and editing, and clips can be extended with further rounds.

How long can a Seedance 2.5 clip be?

ByteDance says a single pass reaches thirty seconds, up from fifteen in 2.0, with the option to extend. Durations run from four to thirty seconds plus an automatic setting.

How is Seedance 2.5 different from 2.0?

Longer single-pass storytelling, timestamp-level editing controls and stronger editing features such as green screen and perspective changes. Seedance 2.5 offers 480p and 720p, while Seedance 2.0 also offers 1080p and 4k.

What output formats does it produce?

Mov and mp4. Mov is useful when the result continues into a post-production tool. Ratios available are 1:1, 3:4, 4:3, 9:16, 16:9, 21:9 and adaptive.

Can it work from reference material?

Yes. Reference-to-video, first and last frame, image-to-video, video extension and video-to-video are all offered, and ByteDance describes precise reference control as a core feature of the model.

How much does Seedance 2.5 cost?

On TokenLab, Seedance 2.5 costs Input $0.00 / Output $10.70 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

Which endpoint should Seedance 2.5 use?

Use https://api.tokenlab.sh/v1/videos/generations for Seedance 2.5. The request example below shows the matching code shape.

Which operations does Seedance 2.5 support?

Seedance 2.5 supports Image to video, Reference to video, Start and end frames to video, Text to video, Video extension, Video to video. Select an operation above to see its endpoint and request example.