Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.dolphy.chat/llms.txt

Use this file to discover all available pages before exploring further.

​
What it does

Takes a still image of a character + a reference video, and produces a new video with your character performing the motion from the reference video.

​
Request

{
  "image_url": "https://your-cdn.example.com/character.jpg",
  "video_url": "https://your-cdn.example.com/reference-motion.mp4",
  "duration_seconds": 5,
  "prompt": "in a futuristic neon city",
  "keep_original_sound": false,
  "character_orientation": "video"
}
FieldNotes
image_urlCharacter still — public URL
video_urlReference motion — public URL
duration_seconds1–30. Affects credit cost (5 credits/sec)
promptOptional scene description
keep_original_soundIf true, preserves the reference video’s audio
character_orientationimage (match the still’s pose) or video (match the reference)

​
Response

{
  "id": "FjkW...",
  "queue_id": "fal-...",
  "model": "kling-v3-motion-control",
  "status": "PROCESSING",
  "credits_charged": 25
}
Poll via POST /v1/video/retrieve with the returned id. Same response shape as regular video retrieve.

​
Billing

5 credits per second of duration_seconds. A 5-second clip = 25 credits. Refunded automatically on failure.