{
  "id": 19155702,
  "title": "Google advances agentic AI custom silicon expanding TPU family with 8tx and 8i Max",
  "url": "https://urgent.news/2026/11/09/google-advances-agentic-ai-custom-silicon-expanding-tpu-family-with",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-09T13:00:14.000Z",
  "source": {
    "name": "SiliconANGLE",
    "slug": "siliconangle",
    "url": "https://siliconangle.com/2026/11/09/google-advances-agentic-ai-custom-silicon-expanding-tpu-family-with-8tx-and-8i-max/"
  },
  "original_language": "en",
  "account": "Google has unveiled two new specialized silicon chips for artificial intelligence as part of its eighth-generation Tensor Processing Unit family. At Google Cloud Next 2026, the company introduced the 8t and 8i TPUs, designed for large-scale training and real-time reasoning, respectively. Today's expansion includes two additional specialized chips: the TPU 8i Max for handling massive context windows and the highest compute density per rack, and the TPU 8tx for secure and confidential inference.\n\nThe TPU 8i Max aims to double the compute, memory, and high-bandwidth memory of its predecessor, TPU 8i, while fitting twice the silicon in the same rack footprint. This allows for more on-chip memory, stronger data movement connections, and reduced latency and power consumption. It is optimized for Google's Gemini frontier model, supporting up to 1 million tokens, or roughly 750,000 words, which is crucial for handling larger models and ultra-long context windows.\n\nMeanwhile, the TPU 8tx is designed for secure and confidential inference, catering to regulated enterprises and governments. It keeps sensitive data and model weights protected within a secure enclave, enabling regulated services and industries to fine-tune models using private data without risking leaks.",
  "summary": "Google LLC today introduced two new specialized silicon chips for artificial intelligence to its eighth-generation Tensor Processing Unit architecture family, aimed at higher compute density and secure inference. At Google Cloud Next 2026 in April, Google introduced two specialized TPUs in the same generation, 8t for large-scale training and 8i designed for real-time reasoning. At the time, […]…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}