News from the AI & ML world
DeeperML
staff@insideAI News
//
Google Cloud has unveiled its seventh-generation Tensor Processing Unit (TPU), named Ironwood. This custom AI accelerator is purpose-built for inference, marking a shift in Google's AI chip development strategy. While previous TPUs handled both training and inference, Ironwood is designed to optimize the deployment of trained AI models for making predictions and generating responses. According to Google, Ironwood will allow for a new "age of inference" where AI agents proactively retrieve and generate data, delivering insights and answers rather than just raw data.
Ironwood boasts impressive technical specifications. When scaled to 9,216 chips per pod, it delivers 42.5 exaflops of computing power. Each chip has a peak compute of 4,614 teraflops, accompanied by 192GB of High Bandwidth Memory. The memory bandwidth reaches 7.2 terabits per second per chip. Google highlights that Ironwood delivers twice the performance per watt compared to its predecessor and is nearly 30 times more power-efficient than Google's first Cloud TPU from 2018.
The focus on inference highlights a pivotal shift in the AI landscape. The industry has seen extensive development of large foundation models, and Ironwood is designed to manage the computational demands of these complex "thinking models," including large language models and Mixture of Experts (MoEs). Its architecture includes a low-latency, high-bandwidth Inter-Chip Interconnect (ICI) network to support coordinated communication at full TPU pod scale. The new TPU scales up to 9,216 liquid-cooled chips. This innovation is aimed at applications requiring real-time processing and predictions, and promises higher intelligence at lower costs.
ImgSrc: insideainews.co
References :
Ironwood boasts impressive technical specifications. When scaled to 9,216 chips per pod, it delivers 42.5 exaflops of computing power. Each chip has a peak compute of 4,614 teraflops, accompanied by 192GB of High Bandwidth Memory. The memory bandwidth reaches 7.2 terabits per second per chip. Google highlights that Ironwood delivers twice the performance per watt compared to its predecessor and is nearly 30 times more power-efficient than Google's first Cloud TPU from 2018.
The focus on inference highlights a pivotal shift in the AI landscape. The industry has seen extensive development of large foundation models, and Ironwood is designed to manage the computational demands of these complex "thinking models," including large language models and Mixture of Experts (MoEs). Its architecture includes a low-latency, high-bandwidth Inter-Chip Interconnect (ICI) network to support coordinated communication at full TPU pod scale. The new TPU scales up to 9,216 liquid-cooled chips. This innovation is aimed at applications requiring real-time processing and predictions, and promises higher intelligence at lower costs.
ImgSrc: insideainews.co
Share:
References :
- insidehpc.com: Google Cloud today introduced its seventh-generation Tensor Processing Unit, "Ironwood," which the company said is it most performant and scalable custom AI accelerator and the first designed specifically for inference.
- www.bigdatawire.com: Google Cloud Preps for Agentic AI Era with ‘Ironwood’ TPU, New Models and Software
- www.nextplatform.com: With “Ironwood†TPU, Google Pushes The AI Accelerator To The Floor
- insideAI News: Google today introduced its seventh-generation Tensor Processing Unit, “Ironwood,†which the company said is it most performant and scalable custom AI accelerator and the first designed specifically for inference.
- venturebeat.com: Google's new Ironwood chip is 24x more powerful than the world's fastest supercomputer.
- BigDATAwire: Google Cloud Preps for Agentic AI Era with ‘Ironwood’ TPU, New Models and Software
- insidehpc.com: Google Cloud today introduced its seventh-generation Tensor Processing Unit, "Ironwood," which the company said is it most performant and scalable custom AI accelerator and the first designed specifically for inference.
- the-decoder.com: Google unveils new AI models, infrastructure, and agent protocol at Cloud Next
- AI News | VentureBeat: Google’s new Agent Development Kit lets enterprises rapidly prototype and deploy AI agents without recoding
- Compute: Introducing Ironwood TPUs and new innovations in AI Hypercomputer
- The Next Platform: With “Ironwood†TPU, Google Pushes The AI Accelerator To The Floor
- Ken Yeung: Google Pushes Agent Interoperability With New Dev Kit and Agent2Agent Standard
- The Tech Basic: Details Google Cloud's New AI Chip.
- insideAI News: Google today introduced its seventh-generation Tensor Processing Unit, "Ironwood," which the company said is it most performant and scalable custom AI accelerator and the first designed specifically for inference.
- venturebeat.com: Google unveils Ironwood TPUs, Gemini 2.5 "thinking models," and Agent2Agent protocol at Cloud Next '25, challenging Microsoft and Amazon with a comprehensive AI strategy that enables multiple AI systems to work together across platforms.
- www.marktechpost.com: Google AI Introduces Ironwood: A Google TPU Purpose-Built for the Age of Inference
- cloud.google.com: Introducing Ironwood TPUs and new innovations in AI Hypercomputer
- Kyle Wiggers ?: Ironwood is Google’s newest AI accelerator chip
Classification:
- HashTags: #AI #GoogleCloud #TPU
- Company: Google
- Target: AI Inference Workloads
- Product: Ironwood TPU
- Feature: Inference
- Type: Research
- Severity: Major