Skip to main content
Decart offers four specialized video models through the AI SDK. Each model is optimized for specific video generation tasks.

​
Available Models

​
lucy-pro-t2v

Text-to-Video Generation The lucy-pro-t2v model generates videos directly from text prompts without requiring an input image.
​
modelId
string
required
lucy-pro-t2v
Capabilities:
  • Text-only input
  • High-quality video generation
  • Supports landscape (16:9) and portrait (9:16) orientations
  • 720p and 480p resolution options
Example: 
import { decart } from '@decartai/ai-sdk-provider';
import { experimental_generateVideo as generateVideo } from 'ai';

const { videos } = await generateVideo({
  model: decart.video('lucy-pro-t2v'),
  prompt: 'A man is riding a horse in a field',
});
Source Reference: src/decart-video-settings.ts:2

​
lucy-pro-i2v

Image-to-Video Generation (Production) The lucy-pro-i2v model animates static images into videos, guided by an optional text prompt.
​
modelId
string
required
lucy-pro-i2v
Capabilities:
  • Requires input image
  • Optional text prompt for animation guidance
  • High-quality production-ready output
  • Supports landscape and portrait orientations
Example: 
const { videos } = await generateVideo({
  model: decart.video('lucy-pro-i2v'),
  prompt: {
    image: imageData,
    text: 'The subject begins to walk forward slowly',
  },
});
Source Reference: examples/tasks/video-generation-i2v.ts:4

​
lucy-dev-i2v

Image-to-Video Generation (Development) The lucy-dev-i2v model provides image-to-video capabilities optimized for development and testing workflows.
​
modelId
string
required
lucy-dev-i2v
Capabilities:
  • Requires input image
  • Optional text prompt for animation guidance
  • Faster processing for development iteration
  • Same API interface as lucy-pro-i2v
Use Cases:
  • Rapid prototyping
  • Testing animation concepts
  • Development environments
Source Reference: src/decart-video-settings.ts:4

​
lucy-motion

Trajectory-Based Motion Control The lucy-motion model enables precise control over camera and subject movement using trajectory points.
​
modelId
string
required
lucy-motion
Capabilities:
  • Requires input image
  • Trajectory-based motion control
  • Precise frame-by-frame positioning
  • Normalized coordinate system (0.0 to 1.0)
Example: 
const { videos } = await generateVideo({
  model: decart.video('lucy-motion'),
  prompt: {
    image: imageData,
    text: 'The subject moves along the specified path',
  },
  providerOptions: {
    decart: {
      trajectory: [
        { frame: 0, x: 0.5, y: 0.5 },
        { frame: 12, x: 0.7, y: 0.9 },
        { frame: 25, x: 0.3, y: 0.1 },
      ],
    },
  },
});
Source References:
  • Model definition: src/decart-video-settings.ts:5
  • Trajectory option: src/decart-video-model.ts:22
  • Example usage: examples/tasks/video-generation-motion.ts:4

​
Model Comparison

Featurelucy-pro-t2vlucy-pro-i2vlucy-dev-i2vlucy-motion
Input TypeText onlyImage + textImage + textImage + text + trajectory
QualityProductionProductionDevelopmentProduction
Motion ControlLimitedLimitedLimitedPrecise
Use CaseText-to-videoImage animationDev/testingMotion control
Trajectory SupportNoNoNoYes

​
Common Settings

All video models support these settings:
​
aspectRatio
string
Video aspect ratio. Supported values: 16:9 (landscape) and 9:16 (portrait).
​
seed
number
Seed value for reproducible results.
​
resolution
string
Video resolution. Supported values: 1280x720 (720p) and 854x480 (480p).
See the Settings page for detailed configuration options.

​
API Specifications

All models implement the AI SDK’s VideoModelV3 specification:
​
specificationVersion
string
v3
​
maxVideosPerCall
number
1 - Each call generates one video
Source Reference: src/decart-video-model.ts:84-86

​
Next Steps

Text-to-Video Guide

Learn how to generate videos from text

Image-to-Video Guide

Animate images into videos

Motion Control

Control movement with trajectories

Settings Reference

Configure video generation