The core innovation in SAM 3.1 is object multiplexing, allowing the model to track up to 16 objects in a single forward pass. Previously, each object required its own dedicated pass, but with multiplexing, SAM 3.1 processes all tracked objects together, eliminating redundant
@aiatmeta
-
SAM 3.1: Object Multiplexing Enhances Video Processing Efficiency
By
–
We’re releasing SAM 3.1: a drop-in update to SAM 3 that introduces object multiplexing to significantly improve video processing efficiency without sacrificing accuracy. We’re sharing this update with the community to help make high-performance applications feasible on smaller,
-
TRIBE v2 Predicts Brain Responses Without Retraining
By
–
Without any retraining, TRIBE v2 can reliably predict the brain responses of individuals it has never seen before, achieving a nearly 2-3x improvement over previous methods for both movies and audiobooks We’re releasing the model, codebase, paper, and demo to help researchers
-
CHMv2 Open Source for Global Carbon and Land Management
By
–
CHMv2 is already supporting public sector efforts in the United States, Europe, and beyond. By making these advances open source, we aim to accelerate research and inform carbon offsetting, reforestation, and land management decisions globally. Read the paper:
-
Canopy Height Maps v2: Open Source Global Forest Mapping Model
By
–
We’re announcing Canopy Height Maps v2 (CHMv2), an open source model for high-resolution global forest canopy mapping, developed in partnership with the @WorldResources.
— AI at Meta (@AIatMeta) 12 mars 2026
CHMv2 leverages our DINOv3 Sat-L vision model, specifically optimized for satellite imagery, to deliver… pic.twitter.com/atTCZgkEsxWe’re announcing Canopy Height Maps v2 (CHMv2), an open source model for high-resolution global forest canopy mapping, developed in partnership with the @WorldResources
. CHMv2 leverages our DINOv3 Sat-L vision model, specifically optimized for satellite imagery, to deliver -
Meta’s Custom Silicon MTIA Powers Next-Generation AI
By
–
Custom silicon is critical to scaling next-gen AI. We’re detailing the evolution of the Meta Training and Inference Accelerator (MTIA), our homegrown silicon family designed to power the next era of AI experiences. Traditional chip cycles span years, but model architectures
-
Meta and AMD Partner on GPU Integration for 6GW Data Center Expansion
By
–
Meta 🤝 AMD
— AI at Meta (@AIatMeta) 24 février 2026
Today we’re announcing a multi-year agreement with @AMD to integrate their latest Instinct GPUs into our global infrastructure. With approximately 6GW of planned data center capacity dedicated to this deployment, we’re scaling our compute capacity to accelerate the… pic.twitter.com/a6lNWsfRciMeta AMD Today we’re announcing a multi-year agreement with @AMD to integrate their latest Instinct GPUs into our global infrastructure. With approximately 6GW of planned data center capacity dedicated to this deployment, we’re scaling our compute capacity to accelerate the
-
Meta’s Personal Superintelligence Vision and India’s AI Impact
By
–
ICYMI: @alexandr_wang spoke at the India AI Impact Summit where he shared Meta’s vision for personal superintelligence and how developers in India are already using AI to solve major societal challenges.
— AI at Meta (@AIatMeta) 19 février 2026
See highlights 👇 and then watch his full speech here:… pic.twitter.com/Ew51XUFQGwICYMI: @alexandr_wang spoke at the India AI Impact Summit where he shared Meta’s vision for personal superintelligence and how developers in India are already using AI to solve major societal challenges. See highlights and then watch his full speech here:
-
Chief AI Officer to Speak at India AI Impact Summit
By
–
Our Chief AI Officer @alexandr_wang will take the stage at the India AI Impact Summit. Date: Thursday, February 19
Time: 1:53pm IST // 12:23am PST Watch the livestream here: -
Meta Team Heads to India for AI Impact Summit and Expo
By
–
Our team is heading to India this week for the AI Impact Summit & Expo Stop by the Meta booth (Exhibition Hall 3, Booth No. 3.7) to meet our team and experience: Demos of research, including Omnilingual Automatic Speech Recognition (ASR) and SeamlessExpressive