TAAFT
Free mode
100% free
Freemium
Free Trial
Deals

SAM 3.1

SAM 3.1 is an updated version of Meta’s Segment Anything Model 3 family, built as a unified image and video segmentation foundation model. It is designed for prompt-based object selection and tracking, supporting multiple prompt types while improving robustness and segmentation quality over prior SAM 3 checkpoints. Meta positions the SAM 3 line around open-vocabulary segmentation, meaning it can identify and segment concept instances beyond fixed label sets, making it useful for editing, tracking, annotation, and broader visual understanding workflows.
New Multimodal Gen 3
Released: March 30, 2026

Overview

SAM 3.1 is Meta’s improved promptable segmentation model for images and video. It supports points, boxes, masks, text, and exemplar prompts, and is designed to segment and track objects more accurately than earlier SAM 3 releases, including open-vocabulary concepts across frames.

About Meta Platforms

We're connecting people to what they care about, powering new, meaningful experiences, and advancing the state-of-the-art through open research and accessible tooling.

Industry: Technology, Information and Media
Company Size: 78.000-79.000
Location: Menlo Park, California, US
Website: ai.meta.com
View Company Profile

Tools using SAM 3.1

No tools found for this model yet.

Last updated: March 30, 2026
0 AIs selected
Clear selection
#
Name
Task