Seedance 2.0: start with Fast, finish with Standard
Seedance 2.0 is ByteDance's multimodal video model for complex motion, multi-shot direction and synchronized audio. First Take Lab currently exposes the verified text-to-video path at 720p and five seconds, with the cost shown before generation.
From 121 credits for five seconds. Fast and Standard passed paid provider canaries on August 12, 2026.
Choose by stage
Fast or Standard?
The cheaper tier should answer whether the shot works. Upgrade only after the action, framing and dialogue are right.
121 credits
720p · 5 seconds · native audio
Use it to validate motion, camera direction and timing at the lower 2.0 cost.
152 credits
720p · 5 seconds · native audio
Use it for the selected take when interaction, physics or final rendering matters more.
What Seedance 2.0 accepts
Text
In integration
The first launch path uses text-to-video.
Image
Next
Image-to-video opens after the text path is stable.
Video
Not yet exposed
Reference video controls are deliberately hidden.
Audio
Not yet exposed
Reference audio waits for multimodal abuse testing.
When 2.0 earns its cost
Complex movement
Sports, dance and physical interactions with several moving parts.
Multi-character shots
Scenes where subjects must react to each other coherently.
Synchronized audio
Dialogue, ambience and effects generated with the video.
Directed camera
Explicit shot changes, tracking motion and cinematic staging.
Same prompt, two real Seedance 2.0 outputs
Prompt: “A small unbranded ceramic automaton sits at a clean wooden workbench and carefully folds one sheet of blue paper into a crane. Medium close-up, one continuous slow clockwise camera orbit, warm morning window light, physically accurate finger and paper movement.”
Seedance 2.0 Fast
1280×720 · 5.04 seconds · audio enabled
121 credits · expected provider cost $1.2095
Completed in 124.9 seconds
Seedance 2.0 Standard
1280×720 · 5.04 seconds · audio enabled
152 credits · expected provider cost $1.5170
Completed in 222.0 seconds
The decision in one table
First-Take method
Spend the expensive take second
- 1Draft the prompt and camera move on the cheapest model that can answer the question.
- 2Run Seedance 2.0 Fast for five seconds to validate interaction and timing.
- 3Change one variable at a time; move to Standard only when the shot is selected.
Learn with evidence, not recipes copied from elsewhere
The paid canary now anchors a reproducible tutorial and prompt library. Each child page owns one intent and links back to this version hub instead of repeating the same article.
Choose by the question you need answered
Does the basic prompt and framing work?
18 credits
Can it stage difficult motion and interaction?
From 121 credits
Do I need the current high-cost flagship?
From 111 credits
Seedance 2.0 FAQ
Can I generate with Seedance 2.0 here now?
Yes. Seedance 2.0 Fast and Standard are available at 720p for five-second text-to-video generations. Both tiers passed paid provider canaries on August 12, 2026.
Is Seedance 2.0 free?
No. The provider charges per generated second. First Take Lab starter credits cover a Seedance 1.0 Lite draft, not a 121-credit Seedance 2.0 Fast run.
Why does First Take Lab start at five seconds and 720p?
A five-second 720p run gives enough evidence to judge motion, framing and timing while limiting the cost of a failed idea. Longer and multimodal combinations stay closed until they have the same production evidence.
Is First Take Lab affiliated with ByteDance?
No. First Take Lab is an independent tool using hosted model APIs through fal.ai. Seedance is developed by ByteDance Seed.