AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative
Topic hub

Meta AI

Meta AI is the AI arm of Meta (Facebook, Instagram, WhatsApp) and the driving force behind the open-weight Llama model family. Its open releases power thousands of derivative models, and its bets on smart glasses and superintelligence labs shape the industry's direction. All our Meta AI coverage lives on this page.

Official site ↗1070 articles
In the glossary:Large Language Model (LLM)Foundation Model
Comparisons
Llama vs MistralDeepSeek vs Llama
Meta identified three trends in Indian e-commerce: AI, video, and sales via chats
Creative·Meta Corporate News

Meta identified three trends in Indian e-commerce: AI, video, and sales via chats

May 22, 2026·2 min
Meta launched an AI assistant in WhatsApp for small businesses in India
Business

Meta launched an AI assistant in WhatsApp for small businesses in India

May 22, 2026·3 min
Alibaba Introduces Its Own AI Chip for Neural Network Training and Inference
Hardware

Alibaba Introduces Its Own AI Chip for Neural Network Training and Inference

May 21, 2026·3 min
Anthropic exposes G20 banks’ vulnerabilities, Meta invests 130 billion in AI
LLM

Anthropic exposes G20 banks’ vulnerabilities, Meta invests 130 billion in AI

May 21, 2026·2 min
Google Reveals Scale: 3.2 Quadrillion AI Tokens Per Month, Sevenfold Growth
Creative·3DNews AI

Google Reveals Scale: 3.2 Quadrillion AI Tokens Per Month, Sevenfold Growth

Google revealed processing 3.2 quadrillion AI tokens per month—seven times more than a year ago—and recommends Gemini 3.5 Flash to reduce costs.

May 21, 2026·3 min
FlashAttention-4: How Together AI Accelerated Attention on Blackwell GPUs
Hardware·Together AI Blog

FlashAttention-4: How Together AI Accelerated Attention on Blackwell GPUs

Together AI unveiled FlashAttention-4, an attention algorithm optimization for Blackwell GPUs that achieves 1605 TFLOPs/s and runs 2.7× faster than Triton.

May 21, 2026·2 min
🎧 Language Models: The Week in Review
Hardware·Hamidun News Podcast

🎧 Language Models: The Week in Review

Meta is laying off 1,000 employees amid record profits to buy silicon chips, while the Vatican prepares a historic tract on protecting the soul from algorithms.

May 21, 2026·15 min
Core42 leased an office in Minneapolis for a 20 MW AI data center
Hardware·TNW

Core42 leased an office in Minneapolis for a 20 MW AI data center

Core42, the cloud and AI infrastructure subsidiary of Abu Dhabi-based Group G42, has signed a lease agreement for a converted office complex in downtown Minneapolis.

May 21, 2026·3 min
Top themes
01ICLR 2026 in Rio: What Researchers Think About Scaling Large ModelsMay 20, 2026·3 min02Mishustin requires public-sector buyers to purchase Sberbank and Yandex AI modelsMay 20, 2026·3 min03YMTC launches IPO on wave of demand for AI chipsMay 20, 2026·3 min04Penn created a hybrid particle for ultra-fast, energy-efficient AIMay 20, 2026·2 min05Meta Invests 200 Billion in Louisiana Data Center: Investors Demand ProofMay 20, 2026·3 min
Young Asian entrepreneurs profit from hateful AI content on Facebook
Creative·Guardian

Young Asian entrepreneurs profit from hateful AI content on Facebook

May 20, 2026·3 min
Meta to move 7,000 employees into AI roles
Creative·3DNews AI

Meta to move 7,000 employees into AI roles

May 20, 2026·3 min
Yandex updated Alice AI's image generator to create images with Russian text
Creative·3DNews AI

Yandex updated Alice AI's image generator to create images with Russian text

May 20, 2026·2 min
Baidu beat forecasts and reaffirmed its bet on AI agents
Autonomous·Bloomberg Tech

Baidu beat forecasts and reaffirmed its bet on AI agents

May 19, 2026·1 min
🎧 Startups and Investments: The Week's Main Stories
Startups·Hamidun News Podcast

🎧 Startups and Investments: The Week's Main Stories

You know, we have a massive pile of fresh analytics, court leaks, and financial reports showing how AI development has become heavy industry requiring capital on…

May 19, 2026·12 min
🎧 Weekly Summary · 2026-W21
Medicine·Hamidun News Podcast

🎧 Weekly Summary · 2026-W21

Imagine this scenario: you feel unwell, start searching online for medical information, find a doctor, describe your symptoms in detail, and receive a diagnosis and…

May 19, 2026·13 min
USA and Philippines Accelerate Development of 4000-Acre AI Hub in Asia
Business·Bloomberg Tech

USA and Philippines Accelerate Development of 4000-Acre AI Hub in Asia

The USA and Philippines are deploying an unprecedented project to build a major AI and manufacturing hub on 4000 acres of land in Southeast Asia.

May 19, 2026·3 min
LetinAR: Nail-sized microlens will become the foundation of AI glasses
Hardware·TechCrunch

LetinAR: Nail-sized microlens will become the foundation of AI glasses

May 19, 2026·2 min
AI Demands Light: NextEra Buys Dominion in Largest Utility Deal
Business·Bloomberg Tech

AI Demands Light: NextEra Buys Dominion in Largest Utility Deal

May 19, 2026·3 min
MAX messenger launched AI Hub: 10 models and 5,900 users in 54 days
Creative·Habr AI

MAX messenger launched AI Hub: 10 models and 5,900 users in 54 days

May 19, 2026·3 min
Meta to lay off 8,000 employees for $145 billion in AI investment
Startups·TNW

Meta to lay off 8,000 employees for $145 billion in AI investment

May 19, 2026·2 min
Revamped Siri in iOS 27 will get an auto-delete option for chats
LLM·The Verge

Revamped Siri in iOS 27 will get an auto-delete option for chats

May 17, 2026·2 min
LEGO Education abandons its own platform and introduces AI and NFC cards
Robotics

LEGO Education abandons its own platform and introduces AI and NFC cards

May 17, 2026·3 min
Meta AI introduces NeuralBench — a framework for testing brain activity models
Research

Meta AI introduces NeuralBench — a framework for testing brain activity models

May 17, 2026·2 min
AI has learned to build itself: Google DeepMind and OpenAI on the path to self-improvement
Robotics

AI has learned to build itself: Google DeepMind and OpenAI on the path to self-improvement

May 17, 2026·3 min
How to compress a language model 3x: a guide to FP8, GPTQ, and SmoothQuant
LLM·MarkTechPost

How to compress a language model 3x: a guide to FP8, GPTQ, and SmoothQuant

Developers received a step-by-step guide to compressing large language models with llmcompressor, comparing the effectiveness of FP8, GPTQ, and SmoothQuant…

May 17, 2026·3 min
Hermes Agent: How to Run a Personal AI Agent on Your Own Server
LLM·Habr AI

Hermes Agent: How to Run a Personal AI Agent on Your Own Server

Nous Research has released Hermes Agent — an open-source AI agent that you can run on your own VPS and train for your specific tasks.

May 17, 2026·2 min