AI · Article
Google Adds Two Specialized AI Chips to Its TPU Lineup
TechCrunch ·

At a glance
Google unveiled two new versions of its in-house AI chip, the TPU. The move aims to reduce its reliance on chips from other companies for its cloud business. One chip focuses on security, while the other doubles memory.
Key points
- TPU 8tx, designed with AMD, is built to train and run AI while keeping data and models protected.
- TPU 8tx can run inference (the process where AI generates answers) alongside training, so it can distill (transfer knowledge from a large model into a smaller one) and train small models on the fly.
- TPU 8i Max packs two TPU 8i inference chips from April into one package, doubling memory, processing power, and bandwidth.
- Both chips extend the eighth generation of TPUs that Google introduced in April.
Why it matters
Microsoft, Amazon, and OpenAI are also building their own AI chips to lessen their dependence on Nvidia, which shows the intensifying race in AI infrastructure.
Original source
Google broadens its AI hardware portfolio with two new specialized chips
TechCrunch · Nov 11, 00:54 KST
Read on TechCrunch ↗AI summarized the original article. Check the original for accuracy. Copyright belongs to TechCrunch.
Other coverage
- The Decoder · Google ships four TPU variants in one year as it ramps up competition with Nvidia
- 중앙일보 · “AI 에이전트 시대, 작업량 수천배” 인프라 특화 나선 구글 [팩플]
- 연합뉴스 · 구글, 신형 AI 칩 'TPU 8i Max' 공개…연산 성능 2배
- 이데일리 · 구글, AI반도체 4종으로 확대…엔비디아 추격 '가속도'
- 조선일보 · 구글, 자체 칩 TPU 기반 제품 2종 추가 공개… 한 해 4종 쏟아냈다
- 한국경제 · 구글, 보안 강화한 새 AI칩 공개…TPU 8tx·TPU 8i MAX 발표
- 연합인포맥스 · 구글, 엔비디아 추격 박차…AI 반도체 'TPU' 제품군 확대
- JTBC · 구글, ‘성능 2배·보안’ 강화한 차세대 AI 칩 공개…‘TPU 8’ 라인업 확대
More from this field
- Anthropic Employee Says Claude Helped Prove Riemann Hypothesis
An Anthropic employee and mathematician, Levent Alpöge, says he used the company's Claude chatbot to create a detailed proof of the Riemann hypothesis over about six weeks. The proof has not yet been checked, and it is over 250,000 lines long in Lean, a computer language used to confirm maths proofs.
BBC News · Nov 21, 08:08 KST - Cole Medin Live-Streams His AI Software Factory Turning Ideas Into Working Software
Cole Medin runs his AI software factory live, taking rough requests from viewers through planning, building, and checking until working software runs on screen. He explains how it works and what he learned building it. The stream also shows a game the system built.
Cole Medin · Nov 21, 08:08 KST - AWS Adds Managed Cluster Extensions for Monitoring and User Management in SageMaker HyperPod Slurm
Amazon SageMaker HyperPod, a service for running Slurm clusters used to build large AI models, now has managed cluster extensions. These let teams manage monitoring and user accounts on running nodes through an API or the console, instead of relying on scripts.
Amazon AWS Machine Learning blog · Nov 21, 08:07 KST - AWS Marketplace Lets AI Agents Buy Software on Their Own
AWS has updated its AWS Marketplace so AI agents can buy software, APIs, data, and content on their own, not just find them. Buyers pay in dollars, and charges are added to their AWS bill.
디지털투데이 · Nov 21, 07:37 KST