{"schema":"postcutoff/event@1","as_of":"2026-11-19T22:11:00+01:00","url":"https://postcutoff.com/e/2026-04-22-google-tpu-8t-8i/","md":"https://postcutoff.com/e/2026-04-22-google-tpu-8t-8i/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-04-22-google-tpu-8t-8i","date":"2026-04-22","date_precision":"day","short_title":"Google unveils eighth-generation TPUs, split into TPU 8t and TPU 8i","deck":null,"takeaway":null,"category":"hardware-compute","category_label":"Chips & compute","importance":3,"confidence":"medium","status":{"key":"partly","labels":["Partly confirmed"]},"sources":[{"n":1,"title":"Google: Our eighth generation TPUs — two chips for the agentic era","url":"https://blog.google/innovation-and-ai/infrastructure-and-cloud/google-cloud/eighth-generation-tpu-agentic-era/","type":"official","group":"primary","domain":"blog.google"},{"n":2,"title":"Google Cloud: TPU 8t and TPU 8i technical deep dive","url":"https://cloud.google.com/blog/products/compute/tpu-8t-and-tpu-8i-technical-deep-dive","type":"official","group":"primary","domain":"cloud.google.com"},{"n":3,"title":"The Next Web: Ironwood launches, eighth-gen split previewed","url":"https://thenextweb.com/news/google-ironwood-tpu-inference-cloud-next","type":"press","group":"press","domain":"thenextweb.com"}],"official":2,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Google"],"title":"Google unveils eighth-generation TPUs, split into TPU 8t (training) and TPU 8i (inference)","summary":"At Google Cloud Next 2026 (April) Google announced its first split TPU generation: TPU 8t for training (pods of 9,600 chips, 2 PB shared memory, 121 exaFLOPS) and TPU 8i for inference (288 GB HBM, 80% better perf/$), both up to 2x better performance-per-watt than Ironwood, which became generally available at the same event.","key_facts":["TPU 8t: ~3x compute per pod vs previous generation; scales to 9,600 chips with 2 PB shared memory; 121 ExaFLOPS; >97% goodput target","TPU 8i: 80% better performance-per-dollar; 288 GB HBM + 384 MB on-chip SRAM; 19.2 Tb/s interconnect for MoE; up to 5x lower on-chip latency","Both: up to 2x performance-per-watt vs Ironwood (TPU v7)","Ironwood (v7) GA: 4.6 PFLOPS per chip, 42.5 EFLOPS per 9,216-chip superpod (press figures)","Press reports: TPU 8t designed with Broadcom and TPU 8i with MediaTek on TSMC 2nm (not confirmed in Google's post)"],"key_numbers":[],"tags":["tpu","chips","compute","inference","training"],"science":null,"body_md":"## What happened\n\nGoogle introduced two purpose-built eighth-generation TPUs at Cloud Next 2026 in Las Vegas, with general availability promised later in 2026 as part of AI Hypercomputer.\n\n## Why it matters\n\nSeparate training and inference silicon reflects how agentic, long-running inference now dominates compute demand, and strengthens Google's position as the main non-NVIDIA accelerator supplier (Anthropic is reported as an anchor customer).","disputed":[],"related":[{"id":"2026-07-22-alphabet-q2-2026-earnings","url":"https://postcutoff.com/e/2026-07-22-alphabet-q2-2026-earnings/","date":"2026-07-22","date_precision":"day","short_title":"Alphabet Q2 2026: Google Cloud +82%, capex guidance raised to up to $205B, Gemini at 22B API tokens/minute","deck":null,"takeaway":"Pichai said Gemini models process 22B API tokens per minute and the Gemini app had 950M MAU.","category":"business","category_label":"Business","importance":3,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":3,"official":1,"filed":"2026-09-29","updated":"2026-09-29","orgs":["Alphabet","Google"]}],"people":[],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-09-29","type":"filed","text":"Created (exact announcement day inferred from press dated 2026-04-22; confidence medium)"}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-09-29","run":null,"sources_read":null,"updated":"2026-09-29","human_review":null,"version":{"date":"2026-10-04"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":null,"in_training_data":true},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":null,"in_training_data":true},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":22,"in_training_data":false},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":null,"in_training_data":true}],"short_url":null}