{"data":{"slug":"google-tpu-v1-v3","etype":"chip","name":"Google TPU v1–v3","cat":"AI Accelerator","subtitle":"The originals: the 2015 inference ASIC that started the custom-silicon era.","summary":"TPU v1 (deployed 2015, revealed 2016) was a 28 nm inference ASIC built in 15 months after Google calculated that voice search at scale would otherwise require doubling its data centers. v2 (2017) added training, HBM and the first pods; v3 (2018) went liquid-cooled at 1,024-chip scale and trained early BERT-era models. Together they proved a hyperscaler could out-build the merchant market for its own workloads — the argument every custom AI chip since has followed. AlphaGo ran on TPUs when it beat Lee Sedol in 2016.","attrs":[{"g":"Overview","l":"Designer","v":"Google"},{"g":"Overview","l":"Deployed","v":"v1 2015 · v2 2017 · v3 2018"},{"b":"Published paper","g":"Compute","l":"TPU v1","v":"92 TOPS INT8, inference only"},{"b":"Vendor spec","g":"Compute","l":"TPU v2","v":"45 TFLOPS BF16 per chip; first 256-chip pods"},{"b":"Vendor spec","g":"Compute","l":"TPU v3","v":"123 TFLOPS BF16 per chip; liquid-cooled 1,024-chip pods"},{"g":"Overview","l":"Legacy","v":"Introduced BF16 to hardware and the pod-scale systems model"}],"kf":{"node":"28 nm (v1)","year":2015,"class":"AI Accelerator","designer":"Google"},"aliases":["TPU","TPU v1","TPU v2","TPU v3"],"ticker":null,"exchange":null,"hq":null,"founded":null,"website":null,"status":"Historical","region":null,"asof":"2026-01-15","sources":[{"url":"https://arxiv.org/abs/1704.04760","label":"TPU v1 paper (ISCA 2017)"}]},"meta":{"found":true,"count":1,"served_from":"site cache (no live upstream calls)","as_of":"2026-10-14T09:58:28+00:00","disclaimer":"Research and information only — not investment advice. Structured records carry basis labels (vendor spec / reported / estimate); market quotes are delayed/indicative.","attribution":"Records compiled from vendor documentation, official filings and reputable public reporting; market quotes via financial data providers. See /data-sources."}}