{"id":15269,"date":"2025-10-04T15:27:19","date_gmt":"2025-10-04T22:27:19","guid":{"rendered":"https:\/\/mattfife.com\/?p=15269"},"modified":"2025-09-19T15:38:54","modified_gmt":"2025-09-19T22:38:54","slug":"google-ironwood-tpu-9216-chips","status":"publish","type":"post","link":"https:\/\/mattfife.com\/?p=15269","title":{"rendered":"Google Ironwood TPU 9216 chips"},"content":{"rendered":"\n

Google\u00a0ended the Hot Chips 2025<\/a> machine learning session with a detailed look at its newest tensor processing unit, Ironwood<\/a>. First revealed at\u00a0Google Cloud Next 25<\/a>\u00a0in April 2025, Ironwood is Google’s first TPU (Tensor Processing Unit) designed primarily for large scale inference workloads – and it’s a whopper.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

The architecture is incredible. It delivers 4,614 TFLOPs of FP8 performance – and eight stacks of HBM3e provide 192GB of memory capacity per chip and is paired with 7.3TB\/s bandwidth. With 1.2TBps of I\/O bandwidth, the system can scale up to 9,216 chips per pod without glue logic and reach a whopping 42.5 exaflops of performance. It absolutely trounces their previous TPUs<\/a>.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

Deployment is already underway at hyperscale in Google Cloud data centers, although the TPU remains an internal platform not available directly to customers.<\/p>\n\n\n\n

Links:<\/p>\n\n\n\n