meta/sam-3
SAM 3 segments anything you can name: give it an image and a word, and it returns the mask.
Model Inputs
Image to segment (.jpg/.png/.webp)
Drag & drop a file here, or click to select
Supported formats: JPG, PNG, WEBP (max 10MB)
Text prompt for segmentation
Apply the mask on the image.
The format of the generated image.
If True, upload and return multiple generated masks as defined by `max_masks`.
Maximum number of masks to return when `return_multiple_masks` is enabled.
Whether to include mask confidence scores.
Whether to include bounding boxes for each mask (when available).
Result
Pricing
meta/
sam-3
Pricing for Synexa AI models works differently from other providers. Instead of being billed by time, you are billed by input and output, making pricing more predictable.
For example, generating 100 requests costs about $0.50.
Check out our docs for more information about how per-request pricing works on Synexa.
| Provider | Price ($) | Saving (%) |
|---|---|---|
| Synexa | $0.005 | - |
| fal | $0.005 | same price |
Readme
SAM 3 is a promptable segmentation foundation model. Name the thing you want isolated in prompt ("wheel", "the red car", "person on the left") and it detects and segments every matching instance.
apply_mask returns the image with the mask composited on; turn it off to get the raw masks. Enable return_multiple_masks with max_masks when several distinct objects match, and include_scores / include_boxes to get confidence values and bounding boxes back.
Point and box prompting exist upstream but are not exposed here, because they take nested coordinate objects.