Close Menu
  • Home
  • Technology
  • Devices
  • Business
  • Science
  • Gadgets
  • Startups
Ayaksız
  • Home
  • Technology

    Lucid and Bolt Plan 25,000 Robotaxis for Europe

    Eylül 18, 2026

    AMD Gives PC Builders More Choice With New Ryzen 5 Desktop Chips

    Eylül 11, 2026

    Apple Expands Device Protection With a $49.99 Family Plan

    Eylül 11, 2026

    Snapdragon 8 Elite Extreme Gen 6 Reportedly Beats MediaTek’s Next Flagship Chipset

    Eylül 11, 2026

    Pinterest CFO Julia Donnelly to Step Down in October

    Ağustos 30, 2026
  • Devices

    Projector Setup: Separate Placement, Lens Shift, Zoom, and Keystone

    Kasım 18, 2026

    Router Mode or Access Point Mode? Give Each Network Device a Clear Job

    Kasım 18, 2026

    Scanner Feeder or Glass? Choose According to the Original

    Kasım 13, 2026

    Borderless Printing: Why the Edges of Your Photo Can Disappear

    Kasım 12, 2026

    Camera AE Lock: Recompose a Scene Without Letting Exposure Drift

    Kasım 11, 2026
  • Business

    Google Pixel 11 Pro Leaked Renders Indicate It Might Sport Thinner Bezels Than the Pixel 10 Pro

    Mart 30, 2026

    Motorola Edge 70 Debuts With Snapdragon 7 Gen 4, Ultra-Slim 5.99mm Body

    Kasım 5, 2025

    Samsung Galaxy S26 Series Could See Price Increase Amid Rising Component Costs: Report

    Kasım 4, 2025

    Amazon’s AWS Partners with NBA to Power New AI Basketball Platform

    Ekim 4, 2025

    Billionaire-Backed Bitcoin Firm OranjeBTC to List on Brazil’s B3 Exchange

    Ekim 4, 2025
  • Science

    Double Rainbows: Why the Outer Arc Reverses Its Colors

    Kasım 18, 2026

    Why Ice Floats—and Why Water Behaves Differently Near Freezing

    Kasım 13, 2026

    Air Pressure and Altitude: Why Two Weather Readings May Not Be Comparable

    Kasım 12, 2026

    A Light-Year Measures Distance: Understanding the Scale Behind Space Headlines

    Kasım 11, 2026

    Why the Sky Is Blue—and Why the Same Atmosphere Produces Red Sunsets

    Kasım 10, 2026
  • Gadgets

    Tripod Center Columns: Extra Height Comes with a Stability Tradeoff

    Kasım 18, 2026

    E Ink Ghosting: Match the Refresh Mode to the Task

    Kasım 13, 2026

    Camera Lens Hoods: Choose the Right Shape and Check the Whole Frame

    Kasım 12, 2026

    Pop Filters and Windscreens: Match Microphone Protection to the Noise

    Kasım 12, 2026

    USB Hub Power: Check Where Your Connected Devices Get Their Energy

    Kasım 11, 2026
  • Startups

    Cache-Control for Startups: No-Cache and No-Store Solve Different Problems

    Kasım 18, 2026

    Dead-Letter Queues: Investigate Before Replaying Failed Messages

    Kasım 13, 2026

    Database Connection Pools: Budget Across Every Application Replica

    Kasım 13, 2026

    Startup Staging Sites: Separate Search Visibility from Access Control

    Kasım 12, 2026

    API Idempotency Keys: Keep a Retry Connected to the Original Operation

    Kasım 11, 2026
Ayaksız
Anasayfa » Blog » Meta Tests Its First In-House AI Training Chip
Tech

Meta Tests Its First In-House AI Training Chip

By ayaksızMart 15, 2025Yorum yapılmamış3 Mins Read
Facebook Twitter Pinterest LinkedIn Tumblr Email
Share
Facebook Twitter LinkedIn Pinterest Email

Meta, the parent company of Facebook, has initiated testing of its first in-house chip designed specifically for training artificial intelligence (AI) systems. This development marks a significant step in Meta’s plan to reduce its reliance on external chip suppliers like Nvidia and move toward producing its own custom silicon. Sources told Reuters that Meta has begun a small deployment of the chip and plans to expand production if the test proves successful.

Meta’s push to develop in-house chips is part of a broader strategy to reduce the high infrastructure costs associated with its AI projects. The company has forecast total 2025 expenses between $114 billion and $119 billion, including up to $65 billion in capital expenditure largely driven by investments in AI infrastructure.

The new chip is a dedicated accelerator, meaning it is built specifically for AI tasks, making it more power-efficient compared to graphics processing units (GPUs) typically used for AI workloads. Meta is collaborating with Taiwan-based TSMC to produce the chip. The initial design, known as the “tape-out,” has been completed, a crucial milestone in chip development. While tape-out is expensive, costing tens of millions of dollars, it is an essential part of the process to test the chip’s functionality.

Meta has experienced setbacks in its Meta Training and Inference Accelerator (MTIA) series in the past, even scrapping one chip after its initial tests failed. However, last year, Meta began using a MTIA inference chip for content recommendation systems on platforms like Facebook and Instagram. This progress has encouraged Meta to pursue further development of custom chips, aiming to use them for both training and inference of AI models, including generative AI products like Meta AI.

Meta plans to start using its own chips by 2026 for training purposes, aiming to reduce costs associated with AI model training. Chris Cox, Meta’s Chief Product Officer, discussed the company’s phased approach, noting that while progress has been slow, the success of the first-generation inference chip for recommendations has been a significant achievement. Despite the setbacks in developing custom chips, Meta continues to rely heavily on Nvidia’s GPUs for its AI needs, making it one of Nvidia’s largest customers.

The broader AI industry has raised questions about the effectiveness of scaling up large language models with ever more data and computing power. Chinese startup DeepSeek has introduced new, more efficient AI models that rely more heavily on inference rather than the computationally expensive training process. This has sparked concerns about the future value of GPUs like those from Nvidia, which have faced significant market volatility this year.

AI Chris Cox DeepSeek Generative AI in-house chips inference Meta Meta AI Nvidia recommendation systems training TSMC
Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
ayaksız
  • Website

Related Posts

Lucid and Bolt Plan 25,000 Robotaxis for Europe

Eylül 18, 2026

AMD Gives PC Builders More Choice With New Ryzen 5 Desktop Chips

Eylül 11, 2026

Apple Expands Device Protection With a $49.99 Family Plan

Eylül 11, 2026
Add A Comment

Comments are closed.

Recent Posts
  • Projector Setup: Separate Placement, Lens Shift, Zoom, and Keystone
  • Cache-Control for Startups: No-Cache and No-Store Solve Different Problems
  • Tripod Center Columns: Extra Height Comes with a Stability Tradeoff
  • Double Rainbows: Why the Outer Arc Reverses Its Colors
  • Router Mode or Access Point Mode? Give Each Network Device a Clear Job
Tags
AI AI Chips AI Infrastructure Amazon Android Apple Artificial Intelligence Bitcoin ChatGPT China Cryptocurrency Cybersecurity data centers DeepSeek Donald Trump Electric Vehicles Elon Musk Ether Generative AI Google Huawei India Investment Iphone Meta Microsoft NASA National Security Nvidia OnePlus OpenAI Oppo Realme Samsung semiconductor industry Social media Sony SpaceX Technology Tesla Tether TikTok Vivo WhatsApp Xiaomi
Ayaksız
Instagram
© 2026 Ayaksız.

Type above and press Enter to search. Press Esc to cancel.