Skip to main content

Wanx AI - Image to Video

Transform static images into dynamic videos using Wanx AI's sophisticated image-to-video generation technology with multiple model variants and LoRA customization options.

Model Information​

  • Model ID: Qubico/wanx
  • Display Name: Wanx AI
  • Provider: Qubico/Wanx AI

API Usage​

API Playground

https://api.1min.ai/api/features

Generated cURL Command:

curl -X POST "https://api.1min.ai/api/features" \
  -H "API-KEY: <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
  "type": "IMAGE_TO_VIDEO",
  "model": "Qubico/wanx",
  "conversationId": "IMAGE_TO_VIDEO",
  "promptObject": {
    "taskType": "img2video-14b",
    "image": "<base64-encoded-image-or-url>",
    "prompt": "A person walking through a peaceful garden",
    "aspectRatio": "16:9",
    "negativePrompt": ""
  }
}'

Parameters​

Required Parameters​

  • taskType (string): Model variant to use for generation
  • image (string): Source image for standard and LoRA models (base64-encoded or URL)
  • imageUrl (string): Start frame image for keyframe model (base64-encoded or URL)
  • aspectRatio (string): Video aspect ratio
  • loraType (string): LoRA effect type (required for LoRA variant)
  • loraStrength (number): LoRA effect strength (required for LoRA variant)

Optional Parameters​

  • prompt (string): Text description to guide the video generation
  • negativePrompt (string): What not to include in the video
  • imageTailUrl (string): End frame image for keyframe model (base64-encoded or URL)

Parameter Details​

taskType​

  • Type: String
  • Required: Yes
  • Values: "img2video-14b", "img2video-14b-lora", "img2video-14b-keyframe"
  • Default: "img2video-14b"
  • Description: Selects the model variant for video generation. Each variant has different capabilities and costs.

image​

  • Type: String
  • Required: Yes (for standard and LoRA models)
  • Description: Source image for video generation. Used by img2video-14b and img2video-14b-lora models.

imageUrl​

  • Type: String
  • Required: Yes (for keyframe model)
  • Description: Start frame image for keyframe model. Only used by img2video-14b-keyframe.

imageTailUrl​

  • Type: String
  • Required: No
  • Description: End frame image for keyframe model. Provides precise control over video progression.

aspectRatio​

  • Type: String
  • Required: Yes
  • Values: "16:9", "9:16", "1:1"
  • Default: "16:9"
  • Description: Output video aspect ratio.

loraType​

  • Type: String
  • Required: Yes (for LoRA model)
  • Values: "rotation-effect", "live-wallpaper", "squish-effect", "cakeify-effect", "pixel-art", "super-saiyan-effect", "inflate-effect", "deflate-effect", "crush-effect", "particalize-effect"
  • Description: Type of LoRA effect to apply. Only available for img2video-14b-lora model.

loraStrength​

  • Type: Number
  • Required: Yes (for LoRA model)
  • Range: 0.1 - 1.0
  • Default: 0.8
  • Description: Strength of the LoRA effect. Higher values produce stronger effects.

prompt​

  • Type: String
  • Required: No
  • Description: Text description to guide the video generation process. While not required, providing a prompt can help achieve more targeted results.

negativePrompt​

  • Type: String
  • Required: No
  • Description: Specify what elements to avoid in the generated video.

Available Model Variants​

Wanx AI offers three specialized image-to-video model variants:

img2video-14b​

  • Parameters: Standard source image, prompt, aspect ratio, negative prompt
  • Use Case: General-purpose image-to-video generation
  • Features: High-quality 14B parameter model for converting static images to videos

img2video-14b-lora​

  • Parameters: Source image, LoRA type, LoRA strength, prompt, aspect ratio, negative prompt
  • Use Case: Stylized video generation with special effects
  • Features: 14B parameter model with LoRA effects like rotation, live wallpaper, pixel art, and various visual transformations
  • Available Effects: 10+ different LoRA types for creative video generation

img2video-14b-keyframe​

  • Parameters: Start frame image, end frame image (optional), prompt, aspect ratio, negative prompt
  • Use Case: Precise video progression control between keyframes
  • Features: Advanced temporal control with start and end frame specification

Response Format​

{}

Example Request​

curl -X POST https://api.1min.ai/api/features \
  -H "API-KEY: your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "type": "IMAGE_TO_VIDEO",
    "model": "Qubico/wanx",
    "conversationId": "IMAGE_TO_VIDEO",
    "promptObject": {
      "taskType": "img2video-14b-lora",
      "image": "data:image/jpeg;base64,/9j/4AAQSkZJRgABAQEAYABgAAD...",
      "prompt": "Gentle wind blowing through tall grass in a meadow at sunset",
      "aspectRatio": "16:9",
      "loraType": "live-wallpaper",
      "loraStrength": 0.8,
      "negativePrompt": "blurry, low quality, static"
    }
  }'

Features​

  • Three specialized model variants with different capabilities
  • High-quality 14-billion parameter models across all variants
  • LoRA support with 10+ creative effects (rotation, pixel art, live wallpaper, etc.)
  • Keyframe control for precise start/end frame video progression
  • Flexible aspect ratio support (16:9, 9:16, 1:1)
  • Negative prompting for content exclusion

Use Cases​

  • Standard Model (img2video-14b): General-purpose video generation from static images
  • LoRA Variant (img2video-14b-lora): Creative video generation with special effects and stylistic transformations
  • Keyframe Variant (img2video-14b-keyframe): Precise video progression control between specific start and end frames
  • Social media content with dynamic effects
  • Marketing materials with controlled transitions
  • Creative projects requiring specific visual styles
  • Animation prototyping with keyframe control