Directed text-to-video
Create a focused campaign, product, portrait, or environmental shot from a detailed motion brief.
Video · Available
Seedance 2.0 multimodal video workflow
Seedance 2.0 AI video generator
Direct Seedance 2.0 video from text, first frames, or supported references with duration, resolution, audio, and output controls.
- Provider
- ByteDance
- Base quote
- 76 credits
Input mode
01 / 03Choose text, first-frame, or reference guidance
Text-to-video allows the model to invent the scene. A first frame anchors opening composition and identity. Reference-guided modes should be used only when the source adds a clear visual, movement, or timing requirement.
Name every source role and protected property. More media can increase ambiguity when images, motion, or audio examples conflict with one another.
Motion controls
02 / 03Set frame, duration, resolution, and audio deliberately
Design the composition for the selected aspect ratio and give one action enough time to finish. Resolution, duration, input media, audio, final-frame output, and other controls can affect the quote.
Request the protected amount after every price-relevant choice is final. Seedance 2, 2.5, Fast, and Mini use separate model entries and should not share an assumed price.
Clip approval
03 / 03Review motion and sound as separate layers
Watch once for subject identity, once for camera, once for environmental continuity, and once for audio. Note the exact time of any geometry change, edge artifact, sudden speed shift, or synchronization problem.
Use the persisted Library file for review. Preserve successful controls and make one targeted change to the prompt or source for the next attempt.
Production-verified output
A real Seedance 2.0 provider acceptance result.
Public brief summary: one controlled continuous shot used to verify provider submission and video output handling; the complete acceptance prompt remains private.
- Verified
- 2026-08-17T12:28:14.222Z
- Public prompt hash
- 5f66491556b5…
- Acceptance settings
- reference-to-video workflow · reference input · provider success · 1 video
Provider task succeeded with one video output during the documented production acceptance. This single production acceptance proves the connected task path, not universal motion quality, continuity, audio quality, or repeatability for every shot.
What Seedance 2.0 does inside CreateForge
Seedance 2.0 is also commonly searched as Seedance 2. CreateForge consolidates those names on one canonical video model page and keeps Fast, Mini, and 2.5 as separate production entries.
Choose the input mode according to the source, then describe subject motion, camera motion, environment, pace, endpoint, and sound as distinct parts of one shot. Configure duration and output controls before quoting.
CreateForge validates input media and persists eligible successful output in the private Library. Review the full clip for identity, continuity, geometry, edge behavior, and audio timing.
Seedance 2.0 use cases
Choose the model for a specific deliverable.
Image-to-video production
Animate an approved first frame while protecting subject identity, geometry, composition, and palette.
Reference-led movement
Use supported images, video, or audio to communicate visual and temporal constraints for a shot.
Generation controls
A consistent form, tuned to the model.
Prompt
requiredDescribe the video to create with Seedance 2.0.
Generation mode
requiredChoose the input workflow before adding reference media.
Text to Video · Frames to Video · Reference to Video
First frame
optionalAccepts up to 1 image file.
Last frame
optionalAccepts up to 1 image file.
Reference images
optionalAccepts up to 9 image files.
Reference videos
optionalAccepts up to 3 video files with a 15 second combined limit.
Reference audio
optionalAccepts up to 3 audio files with a 15 second combined limit.
Aspect ratio
requiredChoose a supported output frame with visual ratio cards.
Adaptive · 16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:9
Resolution
requiredYour selected resolution is included in the credit price.
480P · 720P · 1080P · 4K
Duration
requiredChoose 4–15 seconds in Video Settings.
Prompt blueprint
- 01Write one continuous shot with a start state, action, camera move, pace, and visible ending.
- 02Explain what each image, video, or audio reference contributes and remove conflicting sources.
- 03Enable audio and returned-frame options only when they serve a real downstream requirement.
Product orbit with audio
A matte orange speaker stands centered on a pale stone plinth. It remains still while the camera makes one slow clockwise orbit and a narrow light travels across the surface, product shape and label remain stable, soft room ambience and one subtle mechanical click, no cuts or text.
Separates camera, light, product stability, and sound into reviewable layers.
First-frame portrait motion
Use the supplied portrait as the exact opening identity and composition. The subject looks up, turns slightly toward the window, and holds a calm expression while the camera pushes in gently, face and wardrobe remain stable, soft rain ambience, no dialogue or cut.
Defines one emotional action while protecting identity and source composition.
Related guides
Learn Seedance 2.0 by task.
Video guide
How to use Seedance 2 for directed AI video
Plan a Seedance 2 video from text, an opening frame, or reference media with clear motion, aspect ratio, audio, quote, and result-review decisions.
Read guideWorkflow
Seedance 2 image to video workflow
Use Seedance 2 image to video with a deliberate first frame, bounded motion, continuity rules, current production controls, protected quotes, and saved results.
Read guidePrompt library
Seedance 2 prompts for controlled AI video shots
Twelve Seedance 2 prompt examples for product motion, portraits, environments, image-to-video, camera movement, audio timing, and short narrative shots.
Read guidePricing guide
Seedance 2 pricing for text, image, and reference video
Plan Seedance 2 credits across text, first-frame, and reference workflows with current model facts, parameter-aware quotes, iteration budgets, and settlement.
Read guideSeedance 2.0 FAQ
Know the real product boundary.
Yes in this CreateForge catalog. The Seedance 2.0 alias redirects to this canonical model page.
Yes. Use a supported first-frame mode, state what the image protects, and describe subject movement, camera, pace, and endpoint.
The connected controls expose native audio where supported. Enable it deliberately, describe source and timing, and review synchronization.
The page displays the current minimum. The protected quote for the exact inputs, duration, resolution, and audio choices governs submission.