Apple Will Run Advanced AI Model on Nvidia GPUs via Google Cloud, Ditching Private Infrastructure

Photo of Thomas Richmond
By Thomas Richmond Published

Quick Read

  • Craig Federighi confirmed Apple's FM Cloud Pro runs on Nvidia GPUs inside Google Cloud, abandoning private infrastructure for its most demanding AI workloads.

  • Apple's M-series private cloud ran for two years but couldn't sustain agentic AI at scale, forcing the outsourcing decision.

  • Nvidia data center revenue surged 92% to $75 billion, while Google Cloud grew 63% to $20 billion with a $460 billion backlog.

  • Act now: the analyst who called NVIDIA in 2010 just named his top 10 AI stocks — and Apple didn't make the cut. Grab the names FREE today.

Apple Will Run Advanced AI Model on Nvidia GPUs via Google Cloud, Ditching Private Infrastructure

© 24/7 Wall St.

Apple (NASDAQ:AAPL | AAPL Price Prediction), the most vertically integrated company in tech, is turning to Nvidia and Google Cloud to power its most advanced AI workloads. According to CNBC’s MacKenzie Sigalos, Apple software chief Craig Federighi confirmed that the company’s new frontier language model, FM Cloud Pro, will run on Nvidia GPUs hosted in Google Cloud for agentic workflows and complex reasoning.

Apple Breaks From Its Vertical Integration Playbook

Apple has built FM Cloud Pro to be on par with Google’s Gemini frontier models. As Sigalos reported on CNBC, “They are confirming that they’re going to be using NVIDIA GPUs in the context of Google Cloud to run this Frontier model. That is a big departure for them.”

Apple has spent years engineering an M-series-based private cloud to keep AI compute in-house, consistent with the privacy-first pitch it makes to consumers. Renting frontier compute from a rival hyperscaler runs against that grain.

Apple’s Private AI Infrastructure Hit a Scaling Limit

“Typically, they would want to control that in their private cloud infrastructure. Now reportedly, it just couldn’t sustain this new model. That’s why they’re turning to Google Cloud,” Sigalos said. She added, “They’ve had it up and running for two years. But when you try to scale this agentic experience, they just reportedly couldn’t sustain it.”

Apple’s private cloud, running on M-series Mac chips, continues to operate. What is moving off-premise is specifically the frontier, agentic workload. The lesson for investors is blunt: multi-step agentic reasoning has compute demands that even Apple’s custom silicon could not meet at scale.

Apple’s financial position gives it the flexibility to navigate that trade-off. The company reported Q2 FY26 revenue of $111.184 billion and diluted EPS of $2.01, marking its eighth consecutive EPS beat. Apple also authorized a new $100 billion share repurchase program. Following WWDC 2026, Morgan Stanley raised its Apple price target to $360 from $330, citing a clearer path to AI monetization.

Apple’s Decision Strengthens Nvidia’s Investment Thesis

NVIDIA (NASDAQ:NVDA) shares moved higher on the news. The implicit endorsement: even a company with world-class internal silicon turned to Nvidia for frontier-scale AI.

Nvidia’s most recent quarter underscores why. Data Center revenue hit $75.25 billion, up 92% year over year, and CEO Jensen Huang has framed the moment as “the largest infrastructure expansion in human history.” Nvidia also flagged an expanded collaboration with Google Cloud on agentic and physical AI, the same lane Apple is now plugging into. Nvidia guided next-quarter revenue to $91.0 billion plus or minus 2%.

Alphabet (NASDAQ:GOOGL) is another beneficiary. Google Cloud revenue grew 63% to $20.03 billion in Q1, with backlog nearly doubling quarter over quarter to over $460 billion. Adding Apple as a frontier-AI customer thickens that backlog and reinforces the case for Alphabet’s $180 to $190 billion 2026 capex plan.

The Privacy Tension and What to Watch

Apple’s AI strategy has long been built around privacy and the iOS walled garden. The reported decision to run a frontier AI model on third-party Nvidia GPUs through Google Cloud raises an important question: how does Apple reconcile outsourced compute with that positioning? On-device processing and M-series-powered private cloud infrastructure reportedly remain in place for other workloads, making the distinction important. The frontier and agentic AI layer appears to be the part of the stack that has moved beyond Apple’s own infrastructure.

For investors, the clearest takeaway may be what this means for Nvidia. If Apple ultimately needs Nvidia-powered infrastructure to deliver a competitive frontier AI product, it adds another data point to the thesis that the industry’s most advanced AI workloads continue to rely on Nvidia’s ecosystem. For Apple, the trade-off of using Nvidia’s GPUs seems pretty reasonable: rent the frontier compute, ship the product, and preserve the privacy narrative where possible.

Photo of Thomas Richmond
About the Author Thomas Richmond →

Thomas Richmond is a financial writer and content strategist with 5+ years of experience covering stocks and financial markets. He has published over 250 articles focused on individual stock analysis, helping investors better understand business fundamentals, stock valuations, and long-term opportunities.

Thomas previously served as a Content Lead at TIKR, a stock research platform, where he helped scale the company’s blog to hundreds of articles per month and contributed to a weekly newsletter reaching more than 100,000 investors.

He specializes in breaking down complex companies into clear, actionable insights for everyday investors, with a focus on fundamentals-driven research.

His work has also been featured on platforms including Seeking Alpha and Sure Dividend.

Outside of work, Thomas enjoys weight lifting and soccer.

Continue Reading

Top Gaining Stocks

SJM Vol: 5,284,521
APH Vol: 15,409,690
POOL Vol: 1,587,599
BLDR Vol: 3,106,921
CARR Vol: 10,221,016

Top Losing Stocks

CTRA Vol: 73,319,495
SMCI Vol: 51,854,942
GLW Vol: 16,673,394
NOW Vol: 35,824,740
ENPH Vol: 13,064,701