Fast is a distinct APIXO route with its own exposed modes, resolutions, reference limits, and pricing. Public ByteDance materials do not identify a separate Fast architecture or quantify a speed or quality difference, so those comparisons should not be assumed.
Seedance 2.0 Fast
Seedance 2.0 Fast is APIXO’s speed-oriented route for ByteDance Seedance 2.0, not a separately documented upstream ByteDance model variant. It exposes prompt generation, first-and-last-frame animation, and multimodal reference workflows with optional generated sound and web search, 4–15 second output, and 480p or 720p delivery.
APIXO Modes
3 Workflows
APIXO Price
$0.044–$0.16 / Sec
Resolution
480p / 720p
Duration
4–15 Seconds
Seedance 2.0 launch
February 12, 2026
Create with Seedance 2.0 Fast
Loading workspace...
Direct motion with prompts, keyframes, and mixed references
Multimodal Conditioning
Combine text with image, video, and audio references so a generation can draw from visual composition, motion, camera treatment, effects, or sound-related context.
Keyframe-Constrained Motion
Anchor motion to a supplied first frame or direct a transition between explicit first and last frames without treating them as general-purpose reference images.
Reference-Led Consistency
Carry recognizable subjects, environments, visual treatment, and motion cues into a new sequence. Fidelity remains generative and can weaken when references conflict.
Joint Audiovisual Output
Generate sound with the video so audible events can follow the scene’s timing and visual action. APIXO also permits silent output through its sound control.
Complex Motion Handling
Seedance 2.0 improves physical accuracy and stability in multi-subject interactions and demanding motion scenes compared with Seedance 1.5 Pro.
Camera and Story Control
Interpret prompts and reference footage for camera movement, lighting, performance, transitions, and multi-shot narrative structure within the selected output duration.
Mode-specific production limits
Verified controls available through the Seedance 2.0 Fast route.
3
Generation Modes
4–15 Seconds
Output Duration
6 Ratios + Auto
Aspect Ratios
1–9 Images
Omni Image Inputs
1–3 Videos
Video References
1–3 Audio Files
Audio References
Build short videos from different levels of creative direction
Prompt-Led Scene Exploration
Convert a written brief into short horizontal, vertical, square, or ultrawide concepts with optional sound. This workflow suits early campaign development, storyboard alternatives, visual pitches, and social-content ideation without requiring source media.
Animate Approved Keyframes
Turn a still image into motion or bridge supplied opening and closing frames. Use this for product reveals, storyboard transitions, title sequences, character moments, and design-led clips where endpoints matter more than broad reference mixing.
Combine Campaign References
Provide product shots, environmental images, motion footage, and optional audio cues to direct an advertisement or launch concept. Review branding, identity, object geometry, contact, and reference conflicts before production use.
Translate Movement and Rhythm
Use reference footage and audio as conditioning for camera language, action pacing, performance rhythm, or visual effects in a newly generated clip. The route does not document direct source-video editing or guaranteed audio retention.
Choose the workflow before assembling source media
Each APIXO mode accepts a different reference structure and rejects unsupported combinations.
Start from Text or Keyframes
Use text-to-video when the prompt should define the entire scene. Choose first-and-last-frames when one image must establish the opening or two ordered images must anchor both ends of the generated motion.
Assemble Omni References
For broader conditioning, combine up to nine images, three videos, and three audio files. Each video or audio file must be 2–15 seconds, with no more than 15 seconds per reference-media category.
Configure Output and Delivery
Select 480p or 720p, an integer duration from 4 to 15 seconds, framing, generated sound, and optional web context. Submit asynchronously, then retrieve the resulting MP4 through polling or a callback.
Route identity, billing, and reference constraints
Implementation Notes
ByteDance publicly documents Seedance 2.0, but no separate upstream Seedance 2.0 Fast model or model ID is publicly identified.
First-and-last-frames accepts one or two images; the first is the starting frame and the optional second image is the ending frame.
Omni-reference accepts mixed media, but audio-only requests are rejected; at least one image or video reference is required.
APIXO enables generated sound by default and permits silent output; uploaded audio acts as conditioning rather than documented source-audio preservation.
APIXO documents typical processing around 3–6 minutes for all three workflows; this estimate is variable and not a latency guarantee.
Result URLs may be temporary, so completed videos should be downloaded and stored promptly.
Frequently Asked Questions
Text-to-video, first-and-last-frames, and omni-reference without video input cost $0.08 per output second at 480p or $0.16 per output second at 720p. Images and audio references do not add billable seconds.
At 480p the rate is $0.044 per second; at 720p it is $0.0953. APIXO applies that rate to the generated duration plus the total duration of all reference videos.
APIXO does not document source-audio preservation. Video and audio files provide generation context, while the sound setting controls whether the output includes model-generated audio.
Inspect anatomy, hands, identity, object contact, rapid movement, embedded text, reference fidelity, transitions, dialogue interpretation, pronunciation, and lip synchronization. Complex or conflicting references may require simplification and regeneration.
Explore Other Models
Discover more AI models for your next creative workflow