Alphabet (GOOGL) stock rose 1.7% to $289.50 on Monday after Google added two new chips to its TPU family, sharpening the company's challenge to Nvidia in artificial intelligence hardware. The shares had been up more than 3% in early trading. TPUs, or tensor processing units, are Google's custom-built processors for AI workloads.

Google introduced the TPU 8tx and the TPU 8i Max, both reworked versions of its eighth-generation processors. Bloomberg News first reported the launch earlier in the day.

GOOGLE EXPANDS ITS TPU CHIP LINEUP $GOOGL is introducing two new specialized versions of its eighth-generation TPU systems, bringing its lineup to four configurations, Bloomberg reports. The new systems target two increasingly important AI workloads: 1) TPU 8tx: Designed for… pic.twitter.com/w4prbXR8wl — Wall St Engine (@wallstengine) November 9, 2026

With the additions, Google's eighth-generation TPU lineup now spans four configurations. Think of it as one chip family with two new specialties, each targeting a different customer problem.

The 8tx is built for locked-down data

The TPU 8tx is aimed at government agencies and companies in regulated industries. Google says it is optimized for "secure and confidential inference."

The chip pairs Google's technology with AMD processors and encrypts data at the hardware level. According to the company, it runs up to twice as fast as the previous version, with virtually no trade-off in performance.

Developers can fine-tune models and train AI agents without moving data between separate computing clusters, keeping sensitive information in one place.

The 8i Max goes after the memory wall

The TPU 8i Max targets what the industry calls the "memory wall" — the gap between how fast a chip can calculate and how fast memory can feed it data.

The chip combines two chips in a single package to deliver extra memory and computing power. Google says the design suits coding assistants, mathematical reasoning and drug-discovery tasks.

"TPU 8i Max scales compute and memory in lockstep," Google said. The goal is to keep large workloads from stalling when memory cannot keep pace with computation.

From internal tool to cloud product

Google first showed the original TPU 8 chips earlier this year, with Broadcom (AVGO) helping to design them. The chips were mostly used inside Google, powering services such as Search, YouTube and Gemini. The company has since opened them to cloud customers.

TPU sales helped lift Google Cloud in the September quarter, Alphabet said. Cloud revenue more than doubled to $30.71 billion. That growth is the commercial backdrop for Monday's expansion: the chips have gone from internal infrastructure to a revenue-generating product line.

Oracle (ORCL) will offer TPUs on its cloud platform starting in 2027, making it the first hyperscale cloud outside Google's own to do so. That rollout will put Google-designed silicon on infrastructure it does not operate for the first time — a key milestone to watch.

The Nvidia backdrop

Nvidia (NVDA) has long leaned on developer familiarity with its CUDA software, and its GPUs are known for their general-purpose flexibility. Google has taken the opposite approach, building custom silicon tuned to specific AI workloads rather than one processor for everything — a contrast that now frames the two companies' competition in AI hardware.

Nvidia stock fell 5.8% to $175.83 on Monday. The drop came as the company faced an investigation tied to reports that CEO Jensen Huang had been targeted by a foreign intelligence effort.