Alphabet reported that Google Cloud revenue more than doubled to $30.7 billion in the third quarter of 2026, up 103% year over year, as demand for AI infrastructure, cloud services and its custom Tensor Processing Units accelerated. The October 28 release put total Alphabet revenue at $127.6 billion, a 25% increase.

The Cloud unit earned $11.1 billion in operating profit, compared with $3.6 billion a year earlier. Alphabet said sales of TPU systems contributed meaningfully to Cloud growth. That matters for the TPU-versus-GPU market: Google is no longer positioning its processors solely as internal infrastructure, but increasingly as part of the services and systems customers buy.

During the earnings call, CEO Sundar Pichai said Google was working to bring additional Cloud capacity online. He also reported that the Gemini app had reached one billion users and Google’s first-party model APIs were processing nearly 32 billion tokens per minute. These are company-reported usage figures, not independent measures of model efficiency.

Cloud growth comes with a much larger power bill

Alphabet spent $51.4 billion on capital expenditures during the quarter, largely on AI technical infrastructure, according to management’s earnings commentary. That investment spans servers, networking and data centers, where electrical capacity and cooling remain constraints. Earlier BitcoinVersus.Tech reporting examined the expanding power-and-cooling market and Google’s work on flexible grid capacity.

The financial trade-off was visible elsewhere in the results. Alphabet’s net income fell to $26.2 billion from $35.0 billion a year earlier, while diluted earnings per share declined to $2.12. Non-operating investment losses weighed on profits, even as operating income rose 30% to $40.6 billion. Strong Cloud growth therefore did not translate into an equally strong increase in bottom-line earnings.

What the next quarter must prove

The question is how quickly new infrastructure can turn spending into recurring revenue. TPU availability, facility commissioning, software demand and customer utilization will all shape that answer. The competitive yardstick is also changing: tokens per watt and throughput per megawatt are becoming important alongside raw processor speed. Alphabet’s latest quarter shows both sides of the AI race: record demand for computing capacity and the enormous cost of supplying it.

Editor’s Note: Figures are from Alphabet’s October 28, 2026 earnings materials and management remarks. The featured cover is a cropped Google Cloud photograph of Ironwood TPU infrastructure from the official Google Cloud product announcement; it does not depict an earnings-call event. Support independent technology reporting with a 5,000-satoshi Bitcoin donation by scanning the original QR code below. The donation address is 3C9o19EH5HSiwEPyCTmEKzxhNCbo2X6TTb.

Scan this QR code with a compatible Bitcoin wallet to donate while viewing the article on another screen.

BitcoinVersus.tech is not a financial advisor. This media platform reports on financial subjects purely for informational purposes.