The Fleet: How Google Built AI Factories a Decade Before Everyone Else
Every other case study in this series centres on one campus. Google's does not. It started designing its own AI chips in the mid-2010s, joined them into pods with optical switches that can be rewired in software, trained Gemini across several data centers, and now rents its chips to rivals. This case study is about a fleet, not a site: what a ten-year head start buys, and what it costs.
1. Case file
Google compares an Ironwood pod's 42.5 exaflops with the El Capitan supercomputer's 1.7 exaflops, but that compares FP8 with FP64, which is not like for like. Site estimates come from an Epoch AI-based ranking; Google does not publish them.
2. Timeline
Google's AI factory, 2015 โ 2027
The longest timeline in the series. Solid dots are confirmed by Google or a partner, hollow dots come from the press, โ ๏ธ marks claims we could not confirm and dashed dots are planned. Click a dot for its source.
3. The bet: build the whole stack, early
Google started deploying its first Tensor Processing Unit in its data centers in 2015 and revealed it at Google I/O in 2016. That first chip only ran trained models; from TPU v2 in 2017 onward, TPUs could train them, joined into pods. Google now designs every layer: the chip (with Broadcom as design partner, a deal recently extended to 2031), the optical interconnect, the data center network, the compiler and frameworks (XLA and JAX) and the models (Gemini).
4. Seven generations
Peak compute per chip, by generation
BF16 TFLOPS per chip on a log scale, with the largest pod for each. Toggle to show pod size. Peak numbers are not the same as real training throughput.
Sources: Google 2026 five-generation paper (v2โv5p, Ironwood BF16); Cloud TPU docs (v6e Trillium); Ironwood launch post (FP8). TPU v1 was an inference chip measured in INT8 operations, so it is not on the BF16 scale.
5. The pod and its optical switches
NVIDIA's largest building block, the NVL72, joins 72 GPUs in one rack with copper NVLink. Google's Ironwood pod joins 9,216 chips. Inside each block of chips, TPUs connect directly to their neighbours; between blocks, Google uses optical circuit switches (OCS): mirrors that redirect light from one fiber to another. Google's 2023 TPU v4 paper described how this lets the pod be rewired in software, so it can route around a failed block or carve one pod into slices of different shapes for different jobs, without anyone touching a cable.
Rewire the pod
An illustration with 64 blocks instead of the real number. Click blocks to fail them, then choose a job size: the switch picks healthy blocks and leaves failed ones out.
6. Training across many data centers
Microsoft's Fairwater (Part 3) made headlines in 2025 for linking two sites into one AI superfactory. Google had already done it: its December 2023 Gemini technical report says Gemini Ultra was trained on TPU pods across multiple data centers, which it described as a step up in scale from its earlier flagship model. That is why Google's AI factory is better understood as a fleet of campuses joined by Google's own network than as any single site.
Related reading: AI factory networking and custom silicon vs NVIDIA.
7. Renting chips to rivals
For years TPUs mainly served Google's own products and a few cloud customers. That changed in October 2025, when Google and Anthropic announced that Anthropic would get access to up to one million TPUs, bringing well over a gigawatt of capacity in 2026, in a deal worth tens of billions of dollars. In 2026 Anthropic signed a further agreement with Broadcom and Google for about 3.5 GW of TPUs from 2027, and reports in April 2026 described Google's investment in Anthropic as locking in 5 GW of TPU capacity โ ๏ธ. The Wall Street Journal reported Google funding a $3.2 billion data center in western New York for Anthropic to lease โ ๏ธ.
Who uses whose chips
A simplified map of the labs in this series and the chips they train on, as reported. Tap a lab.
8. Power and emissions
Google spreads its AI capacity across many campuses, buying power from utilities and contracting clean energy, rather than building one giant gas-powered site. But the scale is showing. Alphabet's capital spending was $91.4 billion in 2025, and its 2026 guidance rose from $175โ185 billion in February to $195โ205 billion in July.
Spending and footprint
Google's 2026 Environmental Report, covering 2025, showed electricity use up 37% and total greenhouse gas emissions up 18%, its largest annual rise, driven mainly by manufacturing chips and servers. Operational emissions fell 2% thanks to clean energy contracts, including over 12 GW of new deals, but supply-chain emissions rose 25%, the share of carbon-free electricity on an hourly basis stayed roughly flat, and water use rose 34% to 10.9 billion gallons. Google set a goal in 2020 to run on carbon-free energy every hour of every day by 2030.
9. Six bets compared
| Colossus | Abilene | Fairwater | Rainier | Meta | ||
|---|---|---|---|---|---|---|
| The bet | Speed | Outside capital | Network of sites | Own silicon | Every building type | Whole stack, a decade early |
| Chips | NVIDIA | NVIDIA | NVIDIA | Trainium2 | NVIDIA | TPU v7 Ironwood |
| Unit of scale | 1 campus | 1 campus | 2 linked sites | 1 campus + 1 | Linked buildings + mega-campus | A fleet of campuses |
| Power | Own turbines | Gas microgrid | Utility | Utility GenCo | Own gas / utility gas | Utilities + clean energy contracts |
| Main issue | Permits | Tax breaks | Utility gas | Wells | Stranded plants | Rising emissions and water use |
10. Reading the numbers
- From Google
- Per-chip and pod figures; TPU first deployed 2015, revealed 2016; OCS described in the 2023 TPU v4 paper; Gemini Ultra trained across multiple data centers (Gemini report, Dec 2023); environmental report figures.
- From Alphabet earnings
- 2025 capex $91.4B; 2026 guidance $175โ185B (Feb) โ $195โ205B (Jul), via DCD.
- Deals
- Up to 1M TPUs and well over 1 GW in 2026 (Google and Anthropic, Oct 2025); ~3.5 GW from 2027 with Broadcom (DCD). The 5 GW figure and $3.2B New York data center are press reports โ ๏ธ.
- Estimates
- Site megawatts come from an Epoch AI-based ranking, not Google.
- Illustrations
- The pod rewiring demo uses 64 blocks and a simple allocation rule; the "who uses whose chips" map is simplified.
- Ironwood: the first Google TPU for the age of inference
- Cloud TPU docs: TPU7x (Ironwood)
- TPU v4: An Optically Reconfigurable Supercomputer (2023)
- Google's training supercomputers from TPU v2 to Ironwood (2026)
- Gemini: A Family of Highly Capable Multimodal Models (2023)
- Coverage
- ServeTheHome: Ironwood at Hot Chips 2025
- DCD: Google raises 2026 capex to $195โ205bn
- DCD: Google estimates 2026 capex of up to $185bn
- CNBC: Google and Anthropic cloud deal
- DCD: Broadcom to 2031; Anthropic 3.5 GW of TPUs
- Compute Atlas: Anthropic and Google TPUs, stated vs modelled
- Axios: Google's AI boom sends emissions, power use soaring
- S&P Global: Google's search for power