---
id: "b542"
language: "en"
library: "tools"
official_url: "https://github.com/facebookresearch/segment-anything"
profile_updated_on: "2026-10-20"
---

# Segment Anything (SAM)

The original SAM image-segmentation model extracts masks from point or box prompts, or automatically segments an image.

- Company / organization: Meta AI Research, FAIR
- Company country / region: United States · Company headquarters
- Category: Images & design
- Official website: [https://github.com/facebookresearch/segment-anything](https://github.com/facebookresearch/segment-anything)
- Profile updated: 2026-10-20


## Overview

Segment Anything is Meta's image-segmentation model, using prompts such as points and boxes to identify regions. It produces object masks, and the official code also provides automatic mask generation. Inference code, model weights and example notebooks support integration into custom image-processing workflows.

## Access options

### Model docs · Images & design

The original SAM image-segmentation model extracts masks from point or box prompts, or automatically segments an image.

[Model docs](https://github.com/facebookresearch/segment-anything)

## Official sources and references

- [Official website](https://github.com/facebookresearch/segment-anything)
- [Documentation](https://github.com/facebookresearch/segment-anything/blob/main/notebooks/predictor_example.ipynb)
- [Documentation](https://github.com/facebookresearch/segment-anything/blob/main/notebooks/automatic_mask_generator_example.ipynb)

## Reading & subscriptions

- [Interactive profile](../../../products/segment-anything-sam--b542.html)
- [Read](../../../read/en/products/segment-anything-sam--b542.html)
- [JSON](../../../read/en/products/segment-anything-sam--b542.json)

Descriptions summarize public information; consult official websites for features, pricing and regional availability.