Wireflow is now a Claude connector.

Set it up
Andrew AdamsAndrew Adams · Co-Founder & Operations at Wireflow ·

Seedance 2.5 API

Seedance 2.5 has no confirmed public API yet.

Here is what ByteDance has announced, how access is rolling out, and how to run Seedance video today through one REST endpoint next to Kling 3 and Veo 3.1.

Free to build · no credit card

Seedance 2.1 Text to VideoOpen workflow →
Seedance 2.5 API
Loading interactive canvas…
200+Built on 200+ internal test generations during development
12+12+ AI models benchmarked for optimal output quality
40+40+ configurations tested to find the best defaults

At Wireflow, Andrew and the team have built and iterated on 200+ seedance 2.5 api workflows for creative teams and agencies. The approach below reflects what we've found delivers the most consistent, production-ready results.

01

What Is Known About Seedance 2.5 API Access

ByteDance unveiled Seedance 2.5 in Beijing on June 23, 2026 and opened a public launch window on July 3, 2026, starting as an enterprise beta that rolls out through Dreamina and Jimeng. Nothing shipped between 2.0 and 2.5, so 2.5 is the direct successor to Seedance 2.0. On specs, the model generates 30 seconds in a single pass with no stitching seams, outputs native 4K at 10-bit color depth, accepts up to 50 multimodal reference inputs including images, audio clips, 3D white-box models, and style references, and generates audio in the same latent space as the video.

What has not been announced is the part developers need. There is no published endpoint specification, no confirmed general availability date, and no public pricing for Seedance 2.5, so anyone quoting per-second rates for it today is guessing. For a sense of scale, Seedance 2.0 on fal.ai costs roughly $0.30 per second at 720p with audio and roughly $0.68 per second at 1080p, but that is 2.0 pricing and 4K output over 30 seconds is a different workload. In the meantime, Wireflow runs Seedance 2.0 as one node in a larger pipeline: prototype the workflow on a visual canvas, chain it with image generators, upscalers, and LLM prompt steps, then call the whole pipeline as a single REST endpoint with one Bearer token. When Seedance 2.5 reaches general API access it becomes another node, and your integration code does not change.

02

What You Can Run Through the API Today

01

Text-to-Video Generation

Turn a prompt into a finished clip through a workflow node, with control over aspect ratio, resolution, and motion style.

02

Image-to-Video with Reference Control

Animate a still image or pass a reference image as input to anchor the look, subject, and starting frame of a shot.

03

Video With Audio Attached

Models that generate sound alongside the video return finished clips with audio already in the file, so there is no separate dubbing step.

04

Swap Models Without Code Changes

Kling 3, Veo 3.1, and Seedance sit behind the same request shape. Change the node, not the integration, and add Seedance 2.5 when access opens.

05

Chain with Other Models

Pipe a Nano Banana 2 or Flux 2 Pro image straight into the video node, then upscale the result, all in one workflow.

06

Async Execution, Per-Node Cost

Submit a request, receive an executionId, poll until COMPLETED. Each response itemizes credit usage per node for cost tracking.

More Than Just Seedance 2.5 API

What ByteDance Announced

Seedance 2.5 does 30 seconds in one pass at native 4K and takes up to 50 reference inputs. The Seedance 2.5 page covers the confirmed specs in full.

What ByteDance Announced

Run Seedance 2.0 Today

You do not have to wait to ship video. The Seedance API page covers Seedance 2.0, which runs now through the same REST interface.

Run Seedance 2.0 Today

Benchmark Against Veo 3.1

Veo 3.1 has an official developer API but needs scene-extension chaining past a single clip. The Veo 3.1 video API shares the exact REST interface here, so you can run the same prompt through both.

Benchmark Against Veo 3.1

Built for Programmatic Pipelines

Raw model endpoints leave orchestration to you. Wireflow's programmatic video generation platform adds queues, retries, and cost controls so you ship video at scale from code.

Built for Programmatic Pipelines

Pay Per Generation, Capped

Video API bills spike fast at scale. Wireflow's usage-based AI API pricing charges per execution with spend limits, so a runaway queue cannot blow the budget.

Pay Per Generation, Capped
Open Platform

Build Any AI Workflow

15+

AI Models Integrated

No Watermarks

Full Commercial License

FAQs

Not a confirmed one. ByteDance opened a public launch window on July 3, 2026 as an enterprise beta through Dreamina and Jimeng, but no endpoint specification, general availability date, or public API pricing has been announced.

Andrew Adams

Written by

Andrew Adams · Co-Founder & Operations at Wireflow

Runs client operations and content strategy at Wireflow. Works directly with creative teams and agencies to build production AI workflows.

Content StrategyClient Operations

Run Seedance From Your Codebase

Design the video pipeline visually, then call it as a REST endpoint. Seedance 2.0, Kling 3, and Veo 3.1 behind one API key, with async execution, spend limits, and per-node cost reporting built in. Seedance 2.5 becomes another node once general API access opens.

Free to buildNo credit cardNo GPU or installCancel anytime