Seedance

Seedance is a multimodal AI video model by ByteDance that generates consistent, cinematic videos with strong multi-shot consistency and native audio using text, image, video, and audio references.

v1 Lite T2V

v1 Lite I2V

v1 Pro T2V

v1 Pro I2V

v1 Pro Fast I2V

v1.5 Pro

Seedance 2 Fast

Seedance 2

Configuration

Prompt *

Prompt length: 3-20000 characters.

First Frame (optional)

Click to upload first frame imageJPEG, PNG, WebP, BMP, TIFF, GIF; max 30MB

Last Frame (optional)

Click to upload last frame imageJPEG, PNG, WebP, BMP, TIFF, GIF; max 30MB

Reference Images (optional)

Click to upload reference image(s)JPEG, PNG, WebP, BMP, TIFF, GIF; max 30MB each

Max 9 items. Combined with first/last frame total should not exceed 9 images.

Reference Videos (optional)

Click to upload reference video(s)

Duration: each video [2, 15]s; up to 3 reference videos; combined duration must not exceed 15s.

Reference Audios (optional)

Click to upload reference audio(s)

Duration: each audio [2, 15]s; up to 3 reference audio clips; combined duration must not exceed 15s.

Aspect Ratio

Resolution

Duration

Duration: 5s (4-15s)

Generate audio

Web search (required)

Result Preview

No video generated yet

Fill in the form and click "Generate Video" to start.