Latest Developments in Broadcom’s AI ASIC Business – June 3, 2025

  1. Latest developments in Broadcom’s AI-related ASIC business

  • 3nm Process AI ASIC Chip Mass Production Initiated:

    • Google TPUv6 (the world’s strongest AI XPU) has begun mass production, with projected revenue exceeding$15 billion over its lifecycle.

    • Next Generation TPUv7 Design Completed, set for mass production in 2026; collaboration with Google onTPUv8 (2nm process).

  • Meta’s 3nm MTIA Chip: Mass production in 2026, supporting training and inference tasks.

  • OpenAI with SoftBank/ARM’s 2nm/3nm Chips: Utilizing advanced 3DSOIC packaging, mass production mid-2026.

Latest Developments in Broadcom's AI ASIC Business - June 3, 2025Latest Developments in Broadcom's AI ASIC Business - June 3, 2025

Broadcom’s SoftBank/ARM AI ASIC Program (3DSOIC advanced packaging)

2. Leading Network Chip Technology

  • World’s Highest Performance (102Tbps throughput, 200G SERDES), mass production in the second half of 2025.

  • Supports traditional scale-out and new scale-up scenarios.

  • Tomahawk 6 (3nm) Switch Chip: Broadcom holds over80% market share in the high-performance switching/routing chip market, leading competitors by 1-2 generations.

Latest Developments in Broadcom's AI ASIC Business - June 3, 2025

Related posts

  1. VVVF Door Machine Controller Operation Manual
  2. Renesas Electronics Launches Automotive MCU with Integrated Low-Power Bluetooth
  3. Introduction to Proximity Unlocking Solutions Based on Low Energy Bluetooth (BLE) Technology
  4. Idioms Related to ‘Arm’ and Their Explanations
  5. The Six Core Companies in Intelligent Computing and Chip Manufacturing Equipment
  6. The Social Chip
  7. Phison Electronics Restructures Its Mainland China Operations and Strengthens Cooperation with Shenzhen Hongxin Yu
  8. New Uses for Drones! Collecting Whale ‘Snot’ to Assess Health
  9. Selection Sort Algorithm in C Language
  10. GCC Compiler Toolchain Download and Installation Guide
  11. The Father of C Language Missed His PhD Due to Refusal to Pay Binding Fees, His Thesis Resurfaces 52 Years Later
  12. J. Am. Chem. Soc. | The Team of Tang Fujie and Cheng Jun Reveals the Competitive Effects of Dynamic Solvation Structures in Electrolytes through Machine Learning NMR Spectroscopy
  13. Shocking! Modifying Linux Kernel Code Can Reduce Data Center Energy Consumption by 30%!
  14. Cabin SoC Research: Domestic Production Rate Exceeds 10%, AI-Driven Cabin SoCs to Become Mainstream in the Next 2-3 Years
  15. Understanding the Differences in File Size Reporting: Why `ls` and `du` Show Different Results in Linux
  16. The Evolution of Vehicles: Unveiling the Internet of Vehicles
  17. Meta Testing Its First Self-Developed RISC-V Architecture AI Training Chip
  18. When ChatGPT Meets PLC
  19. The Most Powerful Domestic Alternative Chip Exposed: Huawei’s Secret Weapon is Here! Jensen Huang Publicly Praises Huawei

Leave a Comment