falImage to videoComing soon

fal-ai/live-portrait

Name: fal-ai/live-portrait
Brand: fal

fal-ai/live-portrait is an image to video tool from fal that runs in the background and returns video. Coming soon: listed with its price, not yet open to runs.

This tool is listed but not yet available to run.

Price pending

Pricing

Price pending.

Limits

Runs: In the background
Returns: video
Longest queue wait: 3600s
Run deadline: 3600s
Largest inline input: 1024 KB
Largest input file: 500 MB
Largest output: 4768 MB
Outputs per run: 16

Input

Field	Kind	Default	Description
`video_url` required	URL or file	—	URL of the video to drive the lip syncing.
`image_url` required	URL or file	—	URL of the image to be animated
`blink`	number -30 to 30	0	Amount to blink the eyes
`eyebrow`	number -30 to 30	0	Amount to raise or lower eyebrows
`wink`	number 0 to 25	0	Amount to wink
`pupil_x`	number -45 to 45	0	Amount to move pupils horizontally
`pupil_y`	number -45 to 45	0	Amount to move pupils vertically
`aaa`	number -200 to 200	0	Amount to open mouth in 'aaa' shape
`eee`	number -40 to 40	0	Amount to shape mouth in 'eee' position
`woo`	number -100 to 100	0	Amount to shape mouth in 'woo' position
`smile`	number -2 to 2	0	Amount to smile
`flag_lip_zero`	true or false	true	Whether to set the lip to closed state before animation. Only takes effect when flag_eye_retargeting and flag_lip_retargeting are False.
`rotate_pitch`	number -45 to 45	0	Amount to rotate the face in pitch
`rotate_yaw`	number -45 to 45	0	Amount to rotate the face in yaw
`rotate_roll`	number -45 to 45	0	Amount to rotate the face in roll
`flag_eye_retargeting`	true or false	false	Whether to enable eye retargeting.
`flag_lip_retargeting`	true or false	false	Whether to enable lip retargeting.
`flag_stitching`	true or false	true	Whether to enable stitching. Recommended to set to True.
`flag_relative`	true or false	true	Whether to use relative motion.
`flag_pasteback`	true or false	true	Whether to paste-back/stitch the animated face cropping from the face-cropping space to the original image space.
`flag_do_crop`	true or false	true	Whether to crop the source portrait to the face-cropping space.
`flag_do_rot`	true or false	true	Whether to conduct the rotation when flag_do_crop is True.
`dsize`	number (whole)	512	Size of the output image.
`scale`	number	2.3	Scaling factor for the face crop.
`vx_ratio`	number	0	Horizontal offset ratio for face crop.
`vy_ratio`	number	-0.125	Vertical offset ratio for face crop. Positive values move up, negative values move down.
`batch_size`	number (whole)	32	Batch size for the model. The larger the batch size, the faster the model will run, but the more memory it will consume.

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/live-portrait",
    "input": {
      "video_url": "https://storage.googleapis.com/falserverless/model_tests/live-portrait/liveportrait-example.mp4",
      "image_url": "https://storage.googleapis.com/falserverless/model_tests/live-portrait/XKEmk3mAzGHUjK3qqH-UL.jpeg"
    }
  }'