Futures
Access hundreds of perpetual contracts
TradFi
Gold
One platform for global traditional assets
Options
Hot
Trade European-style vanilla options
Unified Account
Maximize your capital efficiency
Demo Trading
Introduction to Futures Trading
Learn the basics of futures trading
Futures Events
Join events to earn rewards
Demo Trading
Use virtual funds to practice risk-free trading
Launch
CandyDrop
Collect candies to earn airdrops
Launchpool
Quick staking, earn potential new tokens
HODLer Airdrop
Hold GT and get massive airdrops for free
Launchpad
Be early to the next big token project
Alpha Points
Trade on-chain assets and earn airdrops
Futures Points
Earn futures points and claim airdrop rewards
META to Deploy Four New In-House AI Chips by 2027 to Cut Reliance on Nvidia and AMD
Meta Platforms META -0.29% ▼ disclosed plans to deploy four new in‑house AI chips by 2027-end as part of its rapidly growing data center strategy. The chips, MTIA 300, 400, 450, and 500, are the latest additions to the company’s Meta Training and Inference Accelerator (MTIA) program. With these new chips, Meta aims to reduce reliance on third-party suppliers such as Nvidia NVDA +0.55% ▲ and AMD AMD +0.93% ▲ while improving performance and cost efficiency.
Claim 70% Off TipRanks Premium
Unlock hedge fund-level data and powerful investing tools for smarter, sharper decisions
Stay ahead of the market with the latest news and analysis and maximize your portfolio’s potential
Rapid Six**‑Month Chip Release Cycle**
Meta plans to release a new MTIA chip every six months, an unusually fast pace for the chip industry. The first model, MTIA 300, is already in production and powering ranking and recommendation systems across Facebook and Instagram.
Further, Meta says MTIA 400, code‑named Iris, is its first chip to offer cost savings and performance on par with leading commercial products. It has completed lab testing and is moving toward deployment.
The remaining chips, MTIA 450 and 500, named Arke and Astrid, respectively, are slated for mass rollout in 2027, designed to support everything from generative AI inference to large‑scale content recommendations.
Modular Design for Faster Deployment
All four chips use a modular design, which allows Meta to drop new versions into existing server racks without major changes. This makes deployment faster and helps the company scale inference workloads across the hundreds of thousands of chips it already runs.
Vice President of Engineering Yee Jiun Song said the rapid release cycle reflects how quickly Meta’s infrastructure needs are expanding. The company expects to spend $115 billion to $135 billion this year on data center expansion, including its massive Hyperion facility in Louisiana.
Overall, the company still buys tens of billions of dollars’ worth of chips from Nvidia and AMD, but the MTIA program gives Meta more control over supply, pricing, and its long‑term chip strategy.
Is Meta a Good Stock to Buy Now?
Turning to Wall Street, META stock has a Strong Buy consensus rating based on 39 Buy and five Hold recommendations. The average Meta Platforms stock price target of $858.86 implies an upside potential of 32.24%.
Disclaimer & DisclosureReport an Issue