ByteDance
Seedance 2
Seedance 2 is ByteDance's flagship multimodal video family, built on a unified audio-video joint generation architecture that accepts text, image, video and audio — up to 12 reference assets per generation. Six variants cover everything from a cost-effective base tier and original-price Offpeak mode to the full-featured Omni Pro, with six aspect ratios including ultra-wide 21:9.
Seedance 2.0 (released February 2026) introduces audio-video joint generation, native multi-shot composition (4–15s clips), targeted clip/character/action editing and prompt-driven video extension that grows existing footage into continuous shots. The base 2.0 model offers dual Fast/Standard tiers for rapid iteration or quality output, while Seedance 2 Offpeak gives Ultimate users access to lower Sutui pricing. The Omni variants add video-to-video transformation and reference image support for character consistency. Seedance 2.0 Pro is the flagship with top-tier quality across all six aspect ratios, while Omni Pro combines every capability into a single model. The Omni Premium variant provides faster generation for time-sensitive workflows.
Seedance 2 Variants
| Variant | Description |
|---|---|
| Seedance 2.0 | Default model with dual Fast and Standard tiers for text-to-video and image-to-video. |
| Seedance 2 Offpeak | Original-price Seedance tier with the same Fast/Standard workflow and free-tier eligibility. |
| Seedance 2.0 Omni | Adds video-to-video transformation and reference images on top of the base model. |
| Seedance 2.0 Pro | Flagship quality with ultra-wide 21:9 cinematic support and 1080p output. |
| Seedance 2.0 Omni Pro | All capabilities combined: text, image, and video input with references. |
| Seedance 2.0 Omni Premium | Faster generation variant of Omni for time-sensitive production workflows. |
Capabilities
Best For
- Ultra-wide 21:9 cinematic content
- Multimodal references — combine up to 12 text, image, video and audio inputs in one prompt
- Audio-video joint generation with native sync (no separate lipsync pass)
- Targeted edits to specific clips, characters, actions or storylines
- Video extension — grow existing footage into continuous shots
- Multi-format social media production
- Character-consistent video series using references
- Cost-effective rapid prototyping via Fast tier
Strengths
- Unified multimodal architecture (text + image + video + audio inputs in one model)
- Up to 12 reference assets per generation for rich, controllable composition
- Audio-video joint generation with native sync, lip-sync and ambience
- Native multi-shot composition with cinematic 4–15s clips at 480p/720p
- Targeted editing of specified clips/characters/actions and prompt-driven video extension
- Six aspect ratios including 21:9 ultra-wide
- Dual Fast/Standard pricing tiers for budget control
- Omni variants support video-to-video and reference images
- Astorie's default video model for easy onboarding
Limitations
- Native output resolutions are 480p and 720p — use post-process upscaling for 1080p/4K masters
- Base and Pro tiers lack video-to-video support
- Fast tier on the base model is text-to-video only
Tips & Best Practices
Use Seedance 2 on Astorie
Connect Seedance 2 with other AI models on Astorie's infinite canvas. No GPU required — start free.
Get Started Free