Wan 2.2 InfiniteTalk - Ekly API

Generates realistic talking avatar videos from a face image and audio. Ideal for lip-synced speaking videos.


Model id	`wan-2.2-infinitetalk-fal`
Media	video
Submit to	`POST /v1/videos`
Price	13 credits per_second before discounts
Typical time	about 300.0 s

Request

Required: prompt, audio, face_image
Optional: frames_per_second, num_frames, resolution

{
  "model": "wan-2.2-infinitetalk-fal",
  "prompt": "A slow aerial shot over a misty pine forest at dawn",
  "audio": "asset_…",
  "face_image": "asset_…",
  "frames_per_second": 16,
  "num_frames": 80,
  "resolution": "480p"
}

Field	Allowed	Default
`audio`	asset id (up to 1)
`face_image`	asset id (up to 1)
`frames_per_second`	4.0 – 60.0	`16`
`num_frames`	40.0 – 120.0	`80`
`resolution`	`480p`, `580p`, `720p`	`"480p"`

Reference inputs take asset ids from POST /v1/uploads or an earlier output’s asset_id. Prices and allowed values come from the live catalog; GET /v1/models/wan-2.2-infinitetalk-fal is always current.