ModelRushVideoNew
Seedance 2.5 Spicy R2V API
Create a 4–30 second Spicy video with up to 30 images, 10 videos and 10 audio clips. At least one image or video is required. Video input and audio input each have a 30-second total limit. Video input is billed together with output seconds.Combine image, video and audio references480P, 720P, 1080POptional synchronized audio
Starting at
$0.127/ second
Input
async53/30000
Estimated runWith video input, billing includes input and output seconds. This maximum allows up to 30 input seconds; unused credit is released after final usage is known.Current rate: $0.284 Per second of output
Output preview

video720P5sadaptive
Model information
Capabilities and limits
Best for
Private scene creationPersonal video variations
Current customer pricing
For each Operation and price variant, the highest regional customer price is shown.| Model | Capability | Price variant | Customer price |
|---|---|---|---|
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 1080P · input + output seconds | $0.419 / second |
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 1080P | $0.698 / second |
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 480P · input + output seconds | $0.076 / second |
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 480P | $0.127 / second |
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 720P · input + output seconds | $0.171 / second |
| Seedance 2.5 Spicy R2V · modelrush/seedance-2.5-spicy-r2v | Video generation | 720P | $0.284 / second |
Pricing and documentation
Other recommended models

Reference-to-Video
Wan 2.7 R2V
Preserve a subject from image or video references in a new 2–10 second shotStarting pricePricing pending

Text-to-Video
Wan 2.7 Text to Video
Turn a creative brief into a traceable asynchronous video jobStarting pricePricing pending

Text-to-VideoImage-to-VideoReference-to-Video
18+
Wan 3 Spicy
Create cinematic AI videos from text, images, or multimodal referencesStarting pricePricing pending