GET

/v1/schemas/fal/video/fal-ai/sam-3-1/video-rle

Accept:

kind inputactivity videohash 764708b0…extracted Oct 1, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `efd9dc0c822df577cbdffffeeecf5e17a252c0e402b004be552fd747237a7fd8` { "fetchedAt": 1790830869, "extractorVersion": "1", "verifiedAt": null }

Request body

11 properties · 1 required · 2 $defs

property	type	description	constraints
video_url*	string	The URL of the video to be segmented.
apply_mask	boolean	Apply the mask on the video.	default: false
mask_url	string	The URL of the mask to be applied initially.	nullable
frame_index	integer	Frame index used for initial interaction when mask_url is provided.	default: 0
return_object_rles	boolean	Also return every tracked object's own mask in each frame, with an id the object keeps for the whole video. The merged rle is returned as before.	default: false
box_prompts	BoxPrompt[]	List of box prompts with optional frame_index.	default: []
prompt	string	Text prompt for segmentation. Use commas to track multiple objects (e.g., 'person, cloth').	default: ""
boundingbox_zip	boolean	Return per-frame bounding box overlays as a zip archive.	default: false
max_num_objects	integer	Maximum number of objects to track in the video.	1 ≤ n ≤ 128 · default: 16
detection_threshold	number	Detection confidence threshold (0.0-1.0). Lower = more detections but less precise. Defaults: 0.5 for existing, 0.7 for new objects. Try 0.2-0.3 if text prompts fail.	0.01 ≤ n ≤ 1 · default: 0.5
point_prompts	PointPrompt[]	List of point prompts with frame indices.	default: []

Types

2 $defs

BoxPromptobject

object_id	integer	Optional object identifier. Boxes sharing an object id refine the same object.	nullable
x_max	integer	X Max Coordinate of the box	nullable
frame_index	integer	The frame index to interact with.	nullable
y_min	integer	Y Min Coordinate of the box	nullable
x_min	integer	X Min Coordinate of the box	nullable
y_max	integer	Y Max Coordinate of the box	nullable

PointPromptobject

y	integer	Y Coordinate of the prompt	nullable
label	enum	1 for foreground, 0 for background	0 \| 1 · nullable
frame_index	integer	The frame index to interact with.	nullable
x	integer	X Coordinate of the prompt	nullable
object_id	integer	Optional object identifier. Prompts sharing an object id refine the same object. When a text prompt is also given, the id selects which detected object the points refine.	nullable

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"fal","endpointId":"fal-ai/sam-3-1/video-rle","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}