You're looking at a specific version of this model. Jump to the model overview.

sprited /birefnet-video:71dc5698

Input schema

The fields you can use to run this model with an API. If you don’t give a value for a field its default value will be used.

Field Type Default value Description
video
string
Input video (mp4, mov, webm, m4v, or gif).
variant
None
general
Which BiRefNet model to use. 'general' is the all-purpose default; 'toonout' is the anime/stylized fine-tune; the rest are specialized BiRefNet zoo models.
output_format
None
cutout
'cutout' = foreground with background removed; 'mask' = the raw single-channel alpha matte (grayscale video).
video_output_type
None
webm
Output container/codec. 'webm' (VP9, default) and 'mov' (ProRes 4444) keep transparency; 'mp4' (x264) has no alpha so cutouts are composited on background_color; 'gif' is palette-based.
background_color
string
#000000
Background colour (hex) for cutouts in formats without alpha (mp4/gif).
resolution
integer
0

Max: 2048

Inference resolution (square). 0 = the variant's native resolution.
refine_fg
boolean
True
Refine foreground colours (FB blur fusion) to remove background bleed on soft edges. Ignored for 'mask' output.
mask_blur
integer
0

Max: 64

Gaussian blur radius applied to the matte.
mask_offset
integer
0

Min: -64

Max: 64

Grow (+) or shrink (-) the matte by N pixels.
video_quality
None
high
Encoding quality.
video_write_mode
None
balanced
Encoding speed vs size.
precision
None
fp32
GPU inference precision. 'fp32' (default) is full precision; 'fp16' is somewhat faster and lighter with negligible quality difference.
batch_size
integer
4

Min: 1

Max: 16

Frames per GPU forward pass. Higher is faster on long clips but uses more VRAM; auto-reduced to 2 for the HR/2K variants.

Output schema

The shape of the response you’ll get when you run this model with an API.

Schema
{'format': 'uri', 'title': 'Output', 'type': 'string'}