Gibrid.AI
Course · 5 lessons · ~5 minutes

Video builder Seedance 2.0
up to 9 references and audio scenes

ByteDance model with the most flexible scene system: frame transitions, up to 9 reference images, audio-driven scenes, and reference video generation.

⏱ 4–15 sec🖼 up to 9 references🎵 audio scenes🎞 reference-to-video📺 up to 1080p
Practice in Studio →
1
Lesson 1 · 30 seconds

Choosing a version

In the studio, click the model name at the top, then find on the left Seedance — versions on the right:

gibrid.ai/studio · model selection · Seedance
Seedance 2.0NEW
4–15 sec · 1080p · 9 references · audio scenes · reference-to-video · adaptive format
≈ 25 G-tokens / 5 sec
Seedance 2.0 Fast
same, but faster and cheaper · up to 720p
Seedance 1.5 Pro
4 / 8 / 12 sec · 1080p · camera lock (fixed lens)
Seedance 1.0 Lite
5/10 sec · cheapest · start+end frame · seed
12.0 — maximum capabilities; 2.0 Fast — for drafts and iterations.
21.0 Lite — dirt-cheap idea testing with frame-to-frame transitions.
2
Lesson 2 · 1 minute

Mode and parameters

Seedance 2.0 parameters:

parameters panel · Seedance 2.0
Mode
From textFrom imageFrom video reference
Duration
4581015
Format
adaptive16:99:161:121:9
Quality · Audio
480p720p1080p🔊 sound
1«By video reference» (reference) — upload a clip and Seedance will replicate its dynamics and style.
2adaptive — the model will automatically choose the format based on content and references.
3Audio is toggleable (generate_audio): the scene's voiceover is generated together with the video.
3
Lesson 3 · 1 minute

Scenes: frames, references, audio

Seedance works in 'scenes' — and this is important to understand:

"Frames" panel · Seedance 2.0
Scene A · frame-to-frame transition
First frame
Last frame
+ References · up to 9
Ref 1
Ref 2
+add
+add
Scene B · audio-driven (instead of frames)
🎵 audio track
+ reference images
🧩 One-scene rule: start/end frames and the audio track are mutually exclusive. Once you add audio, the start frame automatically becomes a reference and the video follows the audio's rhythm.
🖼 Up to 9 references — character, style, objects: more reference control than any other model in the studio.
4
Lesson 4 · 1 minute

Writing a prompt

Seedance excels at holding composition and loves specific motion descriptions.

✕ Weak
«Dancing to music»
✓ Strong
«A girl dances in a neon hall, her movements in perfect rhythm, the camera smoothly circles around, volumetric light, a light haze, dynamic editing pace»
✅ With an audio scene, write about the CHARACTER of movement ("smoothly", "in rhythm", "sharp accents") — the tempo will come from the track.
✅ References work better when the prompt refers to their essence: "girl from the reference photo", "in the style of the second reference".
5
Lesson 5 · 30 seconds

Generation and result

Hit "Generate" — the task enters the background queue (usually 1–2 minutes). The finished video will appear in your feed and in "My Videos"; tokens for failed generations are refunded automatically.

?
FAQ

FAQ

How to generate video in Seedance 2.0 online?
Open Gibrid Studio (gibrid.ai/studio), select the Seedance 2.0 model in the catalog, set parameters, write your prompt, and click "Generate". Registration takes a minute — new users receive free tokens.

How much does video generation cost in Seedance 2.0?
In Gibrid Studio, generating with Seedance 2.0 costs approximately 25 G-tokens / 5 sec. You only pay for successful generations — tokens for failed ones are returned automatically.

What is the maximum video duration in Seedance 2.0?
Available durations: 2.0: 4–15 s · 1.5 Pro: 4/8/12 · Lite: 5/10. Parameters are set directly in the studio before generation.

Can I use Seedance 2.0 without API keys or VPN?
Yes — Gibrid Studio works in the browser with no API keys, foreign bank cards, or VPN: simply select Seedance 2.0 in the catalog and generate online.

Finale

Cheat sheet

ParameterValuesTip
Duration2.0: 4–15 s · 1.5 Pro: 4/8/12 · Lite: 5/10strict lists — other values will be rounded
Formatadaptive + 6 fixedadaptive = auto-select
Referencesup to 9 imagesstudio's best
Audio scenetrack drives the videoexcludes transition frames
Camerafixed lens (1.5 Pro)static lens without camera shake
Qualityup to 1080p (except 2.0 Fast)Fast — up to 720p

🎼 Clip to music

Audio scene + 3–5 character references = a ready-made musical segment.

🔁 Storytelling

The first and last frames define the story arc — describe the journey between them.

🎞 Style consistency

Reference-to-video: provide a sample clip and get your own story in its dynamic style.

💸 Iterations

Drafts on 2.0 Fast / Lite, final on 2.0 in 1080p.

Generate video in Studio →