Verified tutorial

Facts checked August 12, 2026

How to use Seedance 2.0 without paying Standard prices for an untested idea

This workflow uses the same 720p, five-second settings that passed First Take Lab’s paid Fast and Standard canaries. The objective is not to generate more takes; it is to learn from the cheapest useful take before upgrading.

First run

Fast

121 credits

Locked output

720p · 5s

Audio enabled

Fast canary

124.9s

Provider completion time

Standard canary

222.0s

Provider completion time

Where to access Seedance 2.0

First Take Lab exposes the verified text-to-video endpoints in the browser. You can prepare the prompt before sign-in; the provider call happens only after authentication, safety checks, rate limits and the credit check.

Fast

Use for the first 2.0 attempt and motion validation.

121 credits · 720p · 5s

Open Fast with prompt
Standard

Use for the selected shot when quality matters more than cost.

152 credits · 720p · 5s

Open Standard

The six-step Seedance 2.0 workflow

1. Choose Fast for the first run

Fast costs 121 credits for the locked 720p five-second path. Use Standard only after the shot is selected.

2. Write the prompt in observable parts

State subject and setting, then action, camera movement, timing, light, sound and constraints.

3. Check the charge before submit

Confirm the selected model and exact credit charge. An insufficient balance must stop before the provider call.

4. Generate and watch status

The task moves through queued, processing, success or failed. The paid canaries completed in about 125 seconds for Fast and 222 seconds for Standard.

5. Review one variable at a time

If the action or camera is wrong, change only that part of the prompt before spending on another run.

6. Download or upgrade

Download the result, reuse settings, or switch to Standard for the selected final take.

The prompt structure used in the paid canary

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. Natural room ambience, soft paper folds and a quiet mechanical tick. No dialogue, no captions, no logos, no watermark.

This is a first-party tested prompt, not a copied gallery example. Fast and Standard both returned a 1280×720, 5.04-second MP4 with the requested workshop, automaton and paper-folding action.

Common failure states and what to do next

StateMeaningNext action
Blocked before submitThe prompt or media matched a high-risk rule.Remove the risky request; no provider call should occur.
Insufficient creditsBalance is below the selected model charge.Use 1.0 Lite or get the exact credit gap.
Provider failedThe queued job did not produce a valid video.Confirm the automatic refund, then simplify and retry.
Completed but poorThe inference finished but the creative result missed.Change one prompt variable; completed runs are not refunded.

After the generation completes

  • Play the full clip with audio before downloading.
  • Record the model, seed, prompt and settings for reproducibility.
  • Reuse settings when the shot works; do not rewrite every variable.
  • Move to Standard only when the Fast result has answered the motion and framing questions.

Frequently asked questions

How do I access Seedance 2.0?

Open the First Take Lab generator from this page. Fast and Standard are available for verified five-second 720p text-to-video runs.

Which Seedance 2.0 tier should I choose first?

Choose Fast first. It costs 121 credits versus 152 for Standard and is intended to validate the shot before the final tier.

How long does Seedance 2.0 take?

In the August 12 paid canary, Fast completed in about 125 seconds and Standard in about 222 seconds. Queue time varies, so these are observations rather than guarantees.

Can I use image, audio or video references in this workflow?

Not yet on First Take Lab. Seedance 2.0 supports multimodal inputs at the model level, but this production workflow exposes only the verified text-to-video path.

First Take Lab is independent and not affiliated with ByteDance. Completion times are two dated first-party observations, not latency guarantees.