Google TPU 8

Google TPU 8

Google Cloud’s eighth-generation TPU family announced at Next ’26, split into training-optimized TPU 8t and inference-optimized TPU 8i.

Key facts

  • Type: AI accelerator family
  • TPU 8t: Optimized for training; uses Inter-Chip Interconnect to scale up to 9,600 TPUs and 2 PB of shared high-bandwidth memory in one superpod [src-044]
  • TPU 8i: Optimized for inference; directly connects 1,152 TPUs in one pod and adds 3x more on-chip SRAM to host larger KV caches on-silicon [src-044]
  • Performance claims: TPU 8t delivers 3x Ironwood processing power and up to 2x performance/Watt; TPU 8i delivers 80% better performance per dollar for inference than the prior generation [src-044]
  • Agentic relevance: Google frames TPU 8i as enabling millions of concurrent agents cost-effectively [src-044]

Related entities

Related concepts

Source references

  • [src-044] Thomas Kurian — “Welcome to Google Cloud Next ’26” (2026-04-22)

Robin Cartier perspective

This page is part of Robin Cartier's working AI knowledge graph: a practical research layer for production AI, recommendation systems, experimentation, GEO, and agentic web readiness.

The useful next step is to connect this concept back to applied product leadership and operating models.

Recommended next

Keep reading from this thread

From 491 indexed pages and articles.

  1. Wiki concept LLM Inference Economics The cost and latency structure of serving large language models, driven by compute throughput, memory bandwidth, batch size, context length, KV-cache storage, and Related by tpu
  2. Wiki concept Google Cloud Google's cloud computing business and enterprise AI platform. In [src-044], Thomas Kurian positions Google Cloud around the "Agentic Enterprise": a vertically integrated AI stack Related by google
  3. Insight Agentic Web Readiness A practical checklist for making websites understandable, navigable, and useful for AI agents and answer engines Readers have engaged with this next