falVideo to videoComing soon

fal-ai/void-video-inpainting

Name: fal-ai/void-video-inpainting
Brand: fal

fal-ai/void-video-inpainting is a video 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: 1800s
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 input video containing the object to remove.
`quad_mask_video_url`	text		URL of a mask video for the removal target. For best results this should be a VOID-style quadmask video with 4 grayscale values: 0=object to remove, 63=overlap, 127=affected region, 255=background to keep. A simple binary mask (0=remove, 255=keep) also works. If omitted, the app will generate a temporary mask video from `mask_prompt` using the SAM-3 video endpoint and convert it to a quadmask automatically.
`mask_prompt`	text	—	Text description of what should be masked in the input video, such as the object or person to remove. Used to generate a temporary mask video with SAM-3 when `quad_mask_video_url` is not provided.
`prompt` required	text	—	Text description of the desired background after object removal.
`enable_pass2_refinement`	true or false	false	Run VOID Pass 2 warped-noise refinement after Pass 1. This is slower but can improve temporal consistency on longer clips.
`negative_prompt`	text	The video is not of a high quality, it has a low resolution. Watermark present in each frame. The background is solid. Strange body and strange trajectory. Distortion.	Negative prompt to guide generation away from undesired outputs.
`num_inference_steps`	number (whole) 1 to 50	30	Number of denoising steps. Higher values improve quality but increase latency.
`guidance_scale`	number 0 to 20	1	Classifier-free guidance scale.
`strength`	number 0 to 1	1	Denoising strength. 1.0 means full denoising.
`seed`	number (whole)	—	Random seed for reproducibility.
`num_frames`	number (whole) 1 to 197	85	Temporal window size for inference. The backend snaps this to the nearest CogVideoX-safe value that works with temporal compression and patching. Valid outputs are 69, 77, 85, ..., 197.

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/void-video-inpainting",
    "input": {
      "video_url": "https://storage.googleapis.com/falserverless/kontext-blog/input_video.mp4",
      "prompt": "a video of buildings reflecting on a calm river"
    }
  }'