Image → Videoseedance-2seedance-2-pro-i2v

Seedance 2.0 Pro (I2V)

Animate still images into cinematic video with synchronized audio. Supports start + end frame.

Open in Video Lab Browse all models~187 credits per run

Seedance guideVerified

ByteDance's video lines. Three generations live in our catalog: v1.5 Pro (cinematic motion), Pro/Lite (cost tiers), and Seedance 2.0 (the new sd-2-* family with native audio sync).

Strengths

Tightest scene composition of the major video brands — clean framing, balanced negative space.
Cinematic motion: slow camera reveals, parallax, atmospheric drift look natural.
Seedance 2.0 (sd-2-*) adds native audio/video sync — voices, footsteps, ambient sound.
VIP tier on Seedance 2 unlocks 1080p output and longer durations.
Reference-to-video mode (lite-reference-video, sd-2-omni-reference) blends 1–4 reference images cleanly.

Weaknesses

Lite tier is noticeably softer than Pro — fine for drafts, not finals.
Some sd-2-* variants (Omni Reference No Video) actually output stills, not video — read the model name carefully.
Less explosive action than Kling — Seedance leans contemplative.

Best for

Atmospheric establishing shots, slow reveals, cinematic landscape
Product motion shots — bottles spinning, fabric draping, water pours
Story beats that need controlled motion rather than chaos
Reference-driven character / location consistency

Avoid for

High-energy action — Kling wins there
Text-heavy frames

Prompting tips

Describe the camera move FIRST, then the subject — Seedance's motion engine leads with composition.
Use "slow", "gentle", "glide" verbs for motion — Seedance interprets restraint better than chaos.
Mention specific time of day ("dusk", "golden hour", "twilight") — Seedance's lighting engine handles them well.
For sd-2 (Seedance 2.0): explicit audio cues in the prompt ("footsteps echo", "distant thunder") engage the audio sync.

Parameter tips

camera_fixed: ON for product shots / stable framings. OFF for cinematic motion.
generate_audio (sd-2): ON for finals with ambient sound, OFF for drafts to save cost.
Resolution: 720p is the practical default; 1080p (VIP) on hero shots only.

Parameters

end_image_url
string
The URL of the image to use as the last frame of the video. When provided, the generated video will transition from the starting image to this ending image. Supported formats: JPEG, PNG, WebP. Max 30 MB.
generate_audio
boolean
Whether to generate synchronized audio for the video, including sound effects, ambient sounds, and lip-synced speech. The cost of video generation is the same regardless of whether audio is generated or not.
default: true
duration
string
Duration of the video in seconds. Supports 4 to 15 seconds, or auto to let the model decide based on the prompt.
auto456789101112131415
default: auto
seed
int
Random seed for reproducibility. Note that results may still vary slightly even with the same seed.
image_url
string
The URL of the starting frame image to animate. Supported formats: JPEG, PNG, WebP. Max 30 MB.
resolution
string
Video resolution - 480p for faster generation, 720p for balance, 1080p for highest quality.
480p720p1080p
default: 720p
aspect_ratio
string
The aspect ratio of the generated video. Use 16:9 for landscape, 9:16 for portrait/vertical, 1:1 for square, 21:9 for ultrawide cinematic, or auto to infer from the input image.
auto21:916:94:31:13:49:16
default: auto

You'll need

A text prompt
Start frame
End frameoptional

Try now

More from seedance-2

Seedance 2.0 Pro (T2V)

seedance-2-pro-t2v

Seedance 2.0 Fast (T2V)

seedance-2-fast-t2v

Seedance 2.0 Fast (I2V)

seedance-2-fast-i2v

Seedance 2.0 Pro (Reference)

seedance-2-pro-r2v

Seedance 2.0 Fast (Reference)

seedance-2-fast-r2v