Available nowVideo modelByteDance video generation on SkyVid
Seedance 2.0 AI Video Generator
Create controlled motion from a written scene or multimodal image, video, and audio references. Mention uploaded assets with @ in the prompt, then choose 4–15 seconds and up to 1080p.
Reference-to-video accepts up to 9 images, 3 videos, and 3 audio files. Video and audio references are 2–15 seconds each, with a 15-second total per media type. See ByteDance's model overview.
Why Seedance 2.0
Built for motion that has to hold together
Seedance 2.0 is most useful when the prompt describes a sequence, not just a look. Give the model an action order, a camera plan, and a destination frame.
Complex motion with a clear sequence
Describe multiple connected actions, subject interaction, and changes in shot scale instead of limiting the prompt to a single static moment.
Camera direction in natural language
Control the shot with practical film language such as tracking, orbit, pan, dolly, close-up, and slow motion.
Multimodal reference control
Upload images, videos, and audio, then mention them in the prompt to anchor identity, composition, motion, rhythm, or sound.
Up to 15 seconds in SkyVid
Choose from seven durations between 4 and 15 seconds, six aspect ratios, and three currently enabled resolution tiers.
Workflow
Create a Seedance video in four decisions
The generator keeps the model, controls, and current credit requirement in one place. Configure the shot publicly; sign in only when you continue to generation.
- 01
Choose the starting point
Use text-to-video for a scene built from scratch, or add multimodal references when identity, composition, motion, or sound should come from uploaded material.
- 02
Direct the shot
Write subject action first, then add camera movement, environment motion, lighting, pace, and the intended final frame.
- 03
Set the delivery format
Choose the aspect ratio, resolution, and duration that match the destination rather than generating one oversized master by default.
- 04
Review the exact cost
SkyVid calculates credits from the current configuration before sign-in and carries the draft into the full workspace when you continue.
Prompt patterns
Write motion, not a pile of adjectives
A useful Seedance prompt reads like compact shot direction. State what changes over time and where the shot should finish.
Keep movement readable
“Two fencers circle on a rain-soaked rooftop, exchange three fast strikes, then stop blade-to-blade. Low tracking shot follows their footwork, brief slow motion on impact, city lights reflecting in puddles, end on a tight two-shot.”
Action order + camera path + timing + ending state
Direct light and material
“A brushed-steel watch rotates slowly above dark stone. A narrow light sweeps across the bezel and reveals fine texture, macro camera arcs from the crown to the dial, restrained movement, premium studio contrast, finish on a centered hero frame.”
Object motion + light behavior + macro movement + final composition
Prompt only what should move
“Preserve the face, clothing, and framing. The subject looks up and gives a subtle smile while hair and loose fabric move in a light breeze. Slow dolly in, soft background parallax, natural blinking, settle into the original pose.”
Preservation + subject motion + camera motion + environmental motion
Current SkyVid configuration
Resolution, duration, ratio, and credits
The examples below use an 4-second clip. Values come from the same public model and pricing configuration as the generator, so the visible controls remain the final source of truth.
480p
4s example- Text to video
- 76 credits
- Image to video
- 48 credits
720p
4s example- Text to video
- 164 credits
- Image to video
- 100 credits
1080p
4s example- Text to video
- 408 credits
- Image to video
- 248 credits
Durations
4, 5, 6, 8, 10, 12, 15 seconds
Aspect ratios
16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:9
Reference inputs
Up to 9 images · 3 videos · 3 audio files
Credit requirements can change when provider or product pricing is updated. Always use the live requirement shown in the generator before submitting.
One workspace, more choices
Use Seedance when motion is the brief—switch when it is not
A model page should help you choose, not pretend one model wins every job. These alternatives stay in the same SkyVid workspace and credit system.
Veo 3.1
Try when native audio and cinematic scene construction matter.
View model
Kling 2.5
Compare for flexible general-purpose image animation.
View model
Hailuo
Consider for expressive short clips and fast iteration.
View model
Runway Gen4
Explore when the workflow extends into editing and refinement.
View modelQuestions
Seedance 2.0 FAQ
Answers describe the controls currently available on SkyVid and distinguish them from the native model's broader capability set.
What is Seedance 2.0?
Seedance 2.0 is ByteDance's video generation model for text, image, audio, and video-guided creation. SkyVid supports text-to-video and multimodal reference-to-video with images, videos, and audio.
Can I use Seedance 2.0 for text-to-video and image-to-video?
Yes. Choose Text to video to build a shot entirely from a prompt, or Image to video to upload reference images, videos, or audio and direct how they should influence the subject, camera, environment, and sound.
Which Seedance 2.0 resolutions are available on SkyVid?
SkyVid currently exposes 480p, 720p, 1080p for Seedance 2.0. The available credit cost updates when you change resolution, duration, or input mode.
How long can a Seedance 2.0 video be?
The current SkyVid controls offer 4, 5, 6, 8, 10, 12, 15 second clips. Choose a shorter duration for a focused single action or a longer duration when the shot needs setup and resolution.
How are Seedance 2.0 credits calculated?
Credits are calculated from the selected resolution, duration, and whether the generation starts from text or an image. The generator shows the exact requirement before you continue to the workspace.
Does SkyVid expose Seedance 2.0 audio and video references?
Yes. The Image to video workflow accepts up to 9 reference images, 3 reference videos, and 3 reference audio files. Use @ mentions in the prompt to specify which uploaded asset should control a subject, action, look, motion, or sound.
What are the Seedance 2.0 reference upload limits?
Reference images can have an aspect ratio no greater than 2.5. Videos must be MP4 or MOV with an aspect ratio no greater than 2.5, dimensions from 300 to 6000 pixels, and 409,600 to 2,086,876 total pixels. Audio must be MP3 or WAV. Each video or audio file must be 2–15 seconds, with no more than 15 seconds total for each media type.
What should I include in a Seedance 2.0 prompt?
Describe the subject, action, environment, camera movement, lighting, pacing, and intended ending. For image-to-video, focus on motion and avoid redescribing details that are already fixed by the uploaded image.
Can I use a generated video commercially?
Usage rights depend on the current SkyVid terms, your plan, and the rights you hold in prompts and uploaded materials. Review the Terms and Conditions before publishing generated work commercially.
Ready to direct the first shot?
Create with Seedance 2.0 in SkyVid
Configure the prompt, input mode, resolution, ratio, and duration above. SkyVid shows the credit requirement before you continue.
Try Seedance 2.0 Free