Broadcom: The Emerging NVIDIA Counterbalance

TL;DR

Broadcom's custom silicon division is growing faster than projections, with custom compute revenue up 70% YoY.

Google TPU v6 and Meta MTIA v2 are both Broadcom-designed ASICs that hyperscalers use to reduce NVIDIA dependency for inference.

Marvell is growing in the same space but Broadcom's design win pipeline is 3x larger.

The Full Story

While the press focuses on AMD vs. NVIDIA, the more consequential battle is waged by Broadcom's custom compute division. Every major cloud provider now has an internal chip program: Google TPU, Meta MTIA, ByteDance custom training accelerators — all designed with Broadcom.

Why Broadcom, Not AMD?

AMD competes on the same axis as NVIDIA: general-purpose GPU compute. Broadcom takes a fundamentally different approach — it gives hyperscalers exactly the chip they want, optimized for their specific workloads. No software ecosystem tax, better power efficiency, supply chain diversification.

So What? — Market Impact

If Google, Meta, and ByteDance collectively deploy millions of Broadcom-designed ASICs for inference, NVIDIA's inference market share could erode from ~90% to ~65% by 2028. Training remains NVIDIA's fortress, but inference is where the volume lives.

Sources

The Next Platform: "Broadcom May Become The Biggest Counterbalance To Nvidia"

The Next Platform: "Custom Compute Engine Biz Growing More Than Marvell Ever Hoped"

Broadcom May Become the Biggest Counterbalance to NVIDIA — Custom AI Silicon Is Surging

Executive Summary

📊 Market Strategic Impact