Multi Reference Guide
Seedance 2.5 supports text, image, video, and audio as input references. Combine multiple types of materials to guide AI video generation with greater precision and creative control.
For a detailed Seedance 2.5 tutorial, check out our Full Guide ↗
Prompt Guide
Type @ to quickly insert your uploaded files into the prompt.
Select a file to insert it as a reference (e.g. @Image1, @Video1, @Audio1), followed by a clear description (e.g. "A close-up of @Image1 wearing a jacket") so the AI knows exactly how to use the material.
Material Limits
- Up to 9 reference images (max 30MB each, up to 4K resolution)
- Up to 3 reference videos, total duration ≤ 15 seconds (2s–15s per video)
- Up to 3 reference audios, total duration ≤ 15 seconds (2s–15s per audio)
- Maximum 12 materials in total across all types
Supported Input Combinations
- Text + Image
- Text + Video
- Text + Audio (New)
- Text + Image + Video
- Text + Image + Audio
- Text + Video + Audio
- Text + Image + Video + Audio
Note: To protect personal privacy, real human faces cannot be used as multimodal references (such as image or video references). Please use virtual characters or AI-generated faces instead.
Note: Video and audio duration limits are calculated independently (up to 15s each).
How to Use Seedance 2.0
How to Generate AI Video Online with Seedance 2.0
Three steps from idea to cinematic output — choose your input, write your prompt, and generate in minutes. No learning curve required.
Upload your reference assets
An image locks in visual style. A video clip defines motion. An audio file sets rhythm. You can combine all three — or just start with text.
Describe the scene and tag your assets
Write what you want, then tag what you uploaded: '@Image1 for character style, follow the movement of @Video1.' That's the full instruction set — no special syntax to learn.
Generate, then keep going
Get your video in seconds. Extend it with Continue Filming, edit specific elements, or use the output as a new reference for the next shot.
Core Capabilities
Seedance 2.0 AI Video Generator — six things it does differently.
Each capability maps to a real production problem — built into Seedance 2.0 4K by ByteDance — not bolted on.
Audio-Driven Video Generation
Upload dialogue, music, or narration and Seedance 2.0 generates matching visuals with millisecond-accurate lip-sync. Sound becomes the director — the first AI video generator built this way. Seedance 1.5 Pro introduced this native audio-visual architecture — Seedance 2.0 extends it with 4K output and multimodal inputs.
Persistent Character Identity
Facial features, attire, and body type stay locked across every angle and environment. Your characters remain themselves from frame one to the last — no drift, no inconsistency.
Context-Aware Foley & Ambiance
The model synthesizes synchronized Foley effects — footsteps, rustling fabric — and layers environment-specific ambient sound automatically, producing a full cinematic audio experience.
Narrative Multi-Shot Continuity
Lighting, color grading, and spatial logic stay consistent across multiple shots. Design long-form stories with confidence — Seedance 2.0 handles the continuity so you don't have to.
Multimodal Reference Inputs
Feed images, video clips, and audio simultaneously as creative references. Product photos, past footage, brand music — whatever assets you already have become the blueprint for generation.
Extend & Bridge Any Clip
Generate the next 5–15 seconds of an existing clip without re-describing the scene, or connect two clips with a generated bridge. Same characters, same camera logic — no visible cut. Need stable 1080p batch output at lower credit cost? Seedance 1.0 Pro is built for that workflow.
Gallery
Seedance 2.0 generated video examples
From character-consistent stories to audio-driven scenes — see what creators are generating right now with Seedance 2.0.
Pricing
Start Generating with Seedance 2.0 — No Subscription Required
Pay only for what you generate. Credits never expire, no monthly commitment. Try free before you buy.
Starter
- 900 credits included
- create HD text-to-video or image-to-video clips with natural native audio
- Up to 4K export, No watermark download
- Commercial use license
- Standard queue speed
- Email support
Basic
- 3300 credits included
- faster HD generation for daily content
- Text to Video & Image to Video with native audio
- Up to 4K export, No watermark download
- Commercial use license
- Priority queue speed
- Priority support (email)
Plus
- 5800 credits included
- scale creative runs with better stability and look
- Text to Video & Image to Video with native audio
- Up to 4K export, No watermark download
- Commercial use license
- Faster priority queue + up to 5 concurrent jobs
- Priority support
Professional
- 12200 credits included
- (best value per credit)
- high-volume, professional delivery and teams
- Text to Video & Image to Video with native audio
- Up to 4K export, No watermark download
- Commercial use license
- Fastest queue + up to 10 concurrent jobs
- Full effects pack + early access to new features
- 24/7 priority support
Choose one-time credits • Flexible billing options
Looking for a previous version? Seedance 1.5 Pro · Seedance 1.0 Pro (Legacy)
Help Center
Frequently Asked Questions About Seedance 2.0 Online
How lip-sync works, what inputs are supported, character consistency explained — everything before you start generating.
What is Seedance 2.0 AI Video Generator?
Seedance 2.0 AI Video Generator is ByteDance's multimodal AI video model. It takes text, images, video clips, and audio as inputs — and generates cinematic video with consistent characters and precise lip-sync. Unlike text-only tools, Seedance 2.0 lets you reference images for style, videos for motion, and audio for rhythm simultaneously. You're directing the output rather than guessing at it.
What types of input does Seedance 2.0 AI Video Generator support?
Seedance 2.0 AI Video Generator supports four input types: text prompts, reference images (PNG, JPG, WebP), reference video clips (MP4), and audio files (MP3, WAV). You can upload up to 9 images, 3 video clips, and 3 audio files per generation, and combine them in a single prompt using @commands — for example, "@Image1 for character style, follow the movement of @Video1."
How do I keep my character consistent across different shots?
Upload a reference photo of your character, then tag it in your prompt with @Image1. The Identity Persistence mechanism locks in their face, clothing, and proportions — even across different angles, lighting conditions, and complex movements. No drift, no flickering across shots.
Can I extend or edit a video I've already generated?
Yes. Continue Filming extends any clip forward naturally — same characters, same camera logic, no visible cut. Advanced Editing lets you swap characters, add products, or remove background elements within existing footage without touching anything else in the frame. One shoot, multiple deliverables.
How does lip-sync work in Seedance 2.0 AI Video Generator?
Upload any audio file — voiceover, dialogue, or music. Seedance 2.0 AI Video Generator reads the waveform and drives mouth movement and body rhythm to match the audio's timing and tone. It works for both speech and beat-synced music content. You can also use the original audio track from a reference video clip to drive the visual rhythm.
What video lengths and resolutions does Seedance 2.0 AI Video Generator support?
Generated clips run 4–15 seconds in 720p, 1080p, and 4K. Higher resolutions use more credits per clip. All exports are watermark-free, with a commercial license — you own 100% of what you generate.
How do credits work, and how many do I need per video?
Credits are consumed per generation based on duration and resolution. A standard 5-second clip costs fewer credits than longer or higher-resolution outputs. The Starter plan's 900 credits covers roughly 30–50 standard clips depending on settings chosen. Credits never expire — use them today or six months from now.
Does Seedance 2.0 AI Video Generator require installation or a GPU?
No installation, no GPU, no local setup required. Seedance 2.0 AI Video Generator runs entirely in your browser — any modern device works. Generation happens on the server side, so your hardware specs don't affect output quality or generation speed.
Is there a subscription fee?
No subscription, no monthly charge. You pay once and the credits are yours. Credits never expire — use them at your own pace. Plans start from $9.9 for 900 credits.
Can I use generated videos commercially?
Yes. Every paid plan includes a full commercial license. You retain 100% ownership and copyright of your generated content, allowing you to use it for social media, commercial advertisements, client deliverables, or any other commercial purpose without restrictions.
What if I'm not satisfied with Seedance 2.0 AI Video Generator?
All paid plans come with a 7-day money-back guarantee, no questions asked. Contact support@seedance2pro.net and the refund will be processed promptly.
Does Seedance 2.0 Pro support 4K?
Yes. Seedance 2.0 Pro generates 4K video from images, audio, and text — higher-resolution clips simply use more credits.