Fal icon

Fal

Interact with fal.ai AI model APIs for image, video, speech generation, and LLM text completion

Actions5

Image to Video → Generate Video

AI-generated

Summary

Generate a video from an input image using a selected fal.ai model.

Inputs

  • model (required) — AI model for video generation. Options include Veo 2, Kling v1.6 Pro, OpenAI Sora 2 Pro, Seedance 1.0 Pro Fast, and Veo 3 variants.
  • prompt (required) — Text describing how the image should be animated or the desired motion.
  • imageUrl (required) — URL of the input image (should be 720p or higher).
  • aspectRatio — Aspect ratio for the generated video. Model-dependent; supported ratios include auto, 16:9, 9:16, and others per model.
  • duration — Duration of the generated video in seconds. Model-dependent: Veo 2/3 (5-8s), Kling (5/10s), Sora (4/8/12s), Seedance (3-12s).

Output shape

Returns the fal.ai model response as JSON, containing generated video metadata and URL. Exact structure not declared in provided metadata.

Source dispatches to executeImageToVideo; parameter display and defaults grounded in provided properties.

Discussion