Video

Seedance 2.0

Seedance 2.0 is ByteDance's second-generation multimodal generation model. Use Seedance 2.0 for text-to-video, first-and-last-frames, and omni-reference workflows with per-second billing.

Text/image workflows from $0.091/s · Video-reference from $0.055/s

Loading workspace...

Endpoints

POSThttps://api.apixo.ai/api/v1/generateTask/seedance-2-0Create task

GEThttps://api.apixo.ai/api/v1/statusTask/seedance-2-0?taskId={taskId}Poll status

Authentication

Checking API access...

Sign in to claim free credits, test the Playground, and continue into API integration.

Request body

request_type

async

Use callback with callback_url for webhooks.

provider

official

Matches the Seedance 2.0 manual.

input

object

Mode, prompt, references, size, duration, and audio settings.

Parameters

moderequired

Typeselect

Text-to-video is the simplest starting point. Switch to first-and-last-frames or omni-reference for keyframe or multimodal workflows.text-to-video · omni_reference · first_and_last_frames

promptrequired

Typestring

Describe the subject, action, shot design, lighting, and sound cues you want in the final video.

resolutionrequired

Typeselect

480p for lower cost, 720p for higher quality, or 1080p for highest fidelity.480p · 720p · 1080p

durationrequired

Typeselect

Clip length in seconds. Pricing scales per second.5 · 10 · 15

aspect_ratiooptional

Typeselect

Output frame ratio. Auto lets the model decide.auto · 16:9 · 9:16 · 4:3 · 3:4 · 1:1 · 21:9

soundrequired

Typeboolean

Enable or disable audio generation alongside the video.

web_searchoptional

Typeboolean

Enable real-time web search to reference current information during generation.

Parameter	Type	Required	Description
`mode`	select	required	Text-to-video is the simplest starting point. Switch to first-and-last-frames or omni-reference for keyframe or multimodal workflows.text-to-video · omni_reference · first_and_last_frames
`prompt`	string	required	Describe the subject, action, shot design, lighting, and sound cues you want in the final video.
`resolution`	select	required	480p for lower cost, 720p for higher quality, or 1080p for highest fidelity.480p · 720p · 1080p
`duration`	select	required	Clip length in seconds. Pricing scales per second.5 · 10 · 15
`aspect_ratio`	select	optional	Output frame ratio. Auto lets the model decide.auto · 16:9 · 9:16 · 4:3 · 3:4 · 1:1 · 21:9
`sound`	boolean	required	Enable or disable audio generation alongside the video.
`web_search`	boolean	optional	Enable real-time web search to reference current information during generation.

File parameters (URL arrays)

video_urls[mode=omni_reference]Upload 1–3 reference videos (2–15 s each, total max 15 s). Real-person face videos are supported.Only in mode=omni_reference

audio_urls[mode=omni_reference]Upload 1–3 reference audio files (2–15 s each, total max 15 s).Only in mode=omni_reference

image_urls[mode=first_and_last_frames / omni_reference]Required for first_and_last_frames (1-2 images). Optional for omni_reference (1-9 images).Only in mode=first_and_last_frames / omni_reference

Example (cURL)

curl -X POST "https://api.apixo.ai/api/v1/generateTask/seedance-2-0" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "async",
    "provider": "official",
    "input": {
      "mode": "text-to-video",
      "prompt": "a cinematic tracking shot through a rainy neon street",
      "resolution": "720p",
      "duration": 5,
      "aspect_ratio": "16:9",
      "sound": true,
      "web_search": false
    }
  }'

Full API documentation

Seedance 2.0 API Guide

Seedance 2.0 API for Text-to-Video, First & Last Frame, and Omni Reference Video Generation

Use Seedance 2.0 through APIXO to test Seedance prompts in the playground, compare Seedance 2.0 text-to-video and reference-driven modes, then move the same Seedance workflow into production API calls with per-second pricing.

Rated 4.95/5 by 300 verified APIXO users

Seedance 2.0 workflow

Build and ship Seedance 2.0 API workflows from one page

This Seedance 2.0 landing page is designed for users who want to evaluate Seedance quality first, then keep the same Seedance 2.0 settings for pricing, API key creation, and documentation handoff.

Seedance 2.0 prompt testing

Use the Seedance 2.0 playground to check motion language, camera timing, sound, and composition before moving a Seedance prompt into production.

Seedance 2.0 keyframe control

Use Seedance 2.0 first-and-last-frame mode when a start frame, end frame, or storyboard frame needs to stay visually aligned.

Seedance 2.0 reference control

Use Seedance 2.0 omni-reference mode when image, motion, or audio references should guide identity, rhythm, style, and scene direction.

Seedance 2.0 API handoff

Move from Seedance 2.0 demo preview to API key creation, cURL examples, and production calls without losing the tested prompt settings.

Seedance 2.0 pricing clarity

Compare Seedance 2.0 480p, 720p, and 1080p costs on the same page before choosing a cost profile for batch generation.

Seedance 2.0 production readiness

Use Seedance 2.0 on APIXO when concurrency, real-person portrait support, reference fidelity, and predictable billing matter for a team workflow.

Complete Guide

What is the Seedance 2.0 API?

Seedance 2.0 is ByteDance's multimodal generation model for production workflows. The Seedance 2.0 API supports prompt-only generation, keyframe-driven interpolation, and multimodal reference generation that mixes multiple media inputs.

Compared with first-generation pipelines, Seedance 2.0 improves controllability for camera language, temporal consistency, and cross-modal guidance, making Seedance 2.0 practical for marketing, product storytelling, and creative automation.

Seedance 2.0 Modes and Best-Fit Scenarios

Seedance 2.0 text-to-video: quickest path for script or concept prompts without asset preparation.
Seedance 2.0 first-and-last-frames: best for transition-heavy shots where start/end composition must stay fixed.
Seedance 2.0 omni-reference: best when you need multi-asset control over identity, movement style, and audiovisual mood.

Pricing Logic You Need Before Production

Seedance 2.0 is billed per second, with separate Seedance 2.0 rates by resolution and by whether motion reference is used in omni-reference mode.

Billing Formula

Without motion reference: Seedance 2.0 output seconds × rate. With motion reference: Seedance 2.0 input seconds plus output seconds × rate.

Seedance 2.0 API Technical Specs

Current Seedance 2.0 public API surface for production integration.

Modes

Text-to-video, first-and-last-frames, omni-reference

Inputs

Seedance prompt + image, motion, and audio references

Resolution

480p, 720p, and 1080p

Generation Time

Typically 4–8 minutes

APIXO Exclusive

Capabilities you won't find elsewhere

Exclusive

Unlimited Concurrency

APIXO offers Seedance 2.0 with no concurrency limits, so large Seedance batch jobs can run in parallel without queue bottlenecks.

Exclusive

Real-Person Portrait Support

Seedance 2.0 supports real human portraits and copyrighted material in reference assets, including real-person face references.

Exclusive

Full Capability Access

Unlocks Seedance 2.0 capabilities not exposed by most providers, giving teams fuller control over production pipelines.

What can you build?

Seedance Product Reveals

Use Seedance 2.0 to turn launch copy, brand assets, and visual references into short, polished product reveals with controllable motion language.

Keyframe Animation Pipelines

Use Seedance 2.0 first and last frame inputs to generate smooth transitions for storyboard, previz, and sequence iteration workflows.

Multimodal Creative Production

Blend image, motion, and audio references with Seedance 2.0 to guide style, rhythm, and scene composition together.

Campaign Automation

Connect CMS or workflow tools to create repeatable Seedance 2.0 generation jobs with predictable per-second cost controls.

Notes & limitations

Seedance 2.0 omni-reference mode requires at least one image or motion reference. Audio-only requests fail.
Seedance 2.0 first-and-last-frames mode requires 1–2 reference images.
Each Seedance 2.0 motion or audio reference file must be 2–15 seconds, and total referenced duration cannot exceed 15 seconds each.
Seedance 2.0 generation is long-running: 4–8 minutes is normal, and the first 5 minutes may show no status updates.
Seedance 2.0 Playground offers 5, 10, and 15-second presets; the API supports any integer duration from 4 to 15 seconds.
When Seedance 2.0 omni-reference includes motion URLs, billing uses input seconds plus output seconds times the per-second rate.

Frequently asked questions

Start building

Try the playground above, then move to the API docs when you're ready to integrate.

Try playground View API docs