The artificial intelligence hardware race has evolved well beyond a simple choice between general-purpose GPUs and custom silicon accelerators.
Nvidia (NASDAQ: NVDA) and Broadcom (NASDAQ: AVGO) have each reported earnings, revealing that both companies now look less like chip specialists and more like competing architects for the entire data center.
Nvidia’s data center segment generated $89 billion in revenue during the second quarter, up 117% from the same period a year earlier.
Grace CPUs have generated more than $5 billion in sales over the last year, while the new Vera Rubin CPUs are expected to generate roughly $20 billion in sales by next quarter.
Nvidia’s acquisition of Hugging Face for $12.9 billion underscores how the company is evolving into the platform where generative models are shared, not just the silicon they run on.
Broadcom’s custom accelerators remain its primary growth engine, with XPU shipments comprising 73% of the company’s $16.7 billion in AI semiconductor sales last quarter.
Broadcom’s infrastructure software generated $8.8 billion in revenue, up 29% year over year, showing the VMware acquisition is proving a transformative complement to its semiconductor business.
CEO Hock Tan argues that designing chips around one customer’s model requirements makes them cheaper and more power-efficient than a one-size-fits-all GPU approach.
Nvidia expanded its relationship with Amazon Web Services to include another 2 million GPUs through fiscal 2029, while also helping fund a 12-gigawatt data center campus OpenAI is building in Ohio alongside SoftBank Energy.
Broadcom’s pipeline is similarly massive but far more concentrated, with Anthropic on a path to grow from 1 gigawatt of Ironwood this year to 5 gigawatts of TPU v8i in 2027.
OpenAI has 1.3 gigawatts of Jalapeno inference chips deployed in 2027 and more than 5 gigawatts planned for 2028, while Meta has three generations of custom MTIA chips booked through 2027.
Nvidia guided its current-quarter revenue to $108 billion, with CFO Colette Kress describing it as a “supply-constrained outlook” and noting that unconstrained demand would push growth even higher.
Broadcom guided its current quarter to approximately $34.8 billion in revenue, with AI semiconductors growing 236% year over year to $21.7 billion.
Broadcom’s AI revenue is expected to hit $115 billion in fiscal 2027 and surge to $230 billion by fiscal 2028, effectively doubling for two consecutive years.
Despite that impressive growth trajectory, valuation tells a telling story, with Nvidia trading around 28 times trailing earnings versus Broadcom’s approximately 46 times trailing price-to-earnings ratio.
Nvidia’s broader customer base, comprehensive software platform, and stronger balance sheet make it the more compelling buy despite Broadcom’s faster near-term AI revenue growth.
