Latest News
Terminal‑Bench 2.0 launches with Harbor, testing any container‑installable agent Teams tackle new prompt injection attacks, boost model mitigations ChatLLM Review: All‑in‑One AI Suits Small Teams, Startups and Freelancers Notion 3.0 uses GPT‑5 to let AI run autonomous stakeholder feedback workflows Interactive AI Agent Uses OpenAI Function Schemas for Rapid ML Tasks Huang says AI is industry backbone, while panelists like Li voice skepticism ComputeEval 2025.2 expands to 232 CUDA challenges, upping LLM test difficulty Firebase Studio uses Code OSS editor to simplify full‑stack app development Terminal‑Bench 2.0 launches with Harbor, testing any container‑installable agent Teams tackle new prompt injection attacks, boost model mitigations ChatLLM Review: All‑in‑One AI Suits Small Teams, Startups and Freelancers Notion 3.0 uses GPT‑5 to let AI run autonomous stakeholder feedback workflows Interactive AI Agent Uses OpenAI Function Schemas for Rapid ML Tasks Huang says AI is industry backbone, while panelists like Li voice skepticism ComputeEval 2025.2 expands to 232 CUDA challenges, upping LLM test difficulty Firebase Studio uses Code OSS editor to simplify full‑stack app development
Our content generation service is experiencing issues. A human-curated summary is being prepared.
Research & Benchmarks

Google's Ironwood TPU to be generally available on Cloud in weeks

By
• 2 min read

Google is about to open the doors on its newest AI‑focused accelerator, the Ironwood TPU, for anyone with a Google Cloud account. The rollout is slated for “the coming weeks,” meaning developers and enterprises can start provisioning the hardware without a lengthy waitlist. While the chip itself is a product of Google’s internal silicon team, its relevance stretches beyond a single service.

Companies that rely on massive compute—think large‑scale analytics firms or studios training next‑gen models—have already been tapping Google’s custom processors. The move also signals that the same silicon powering Google’s own research will soon be a standard offering on the public cloud. In short, the hardware that underpins projects like Gemini, Imagen and Veo is about to become a shared resource.

TPUs are chips that are specifically designed to handle AI workloads. Besides providing it for customers on Google Cloud, the company also uses it to train and deploy the Gemini, Imagen, Veo and other families of its AI models. Additionally, large‑scale Google Cloud customers have also utilised TPUs.

TPUs are chips that are specifically designed to handle AI workloads. Besides providing it for customers on Google Cloud, the company also uses it to train and deploy the Gemini, Imagen, Veo and other families of its AI models. Additionally, large-scale Google Cloud customers have also utilised TPUs for their AI workloads.

Anthropic, the company behind the Claude family of AI models, has long utilised TPUs via Google Cloud for its workloads and has recently expanded its partnership with Google to deploy over 1 million new TPUs. Indian multinational conglomerate Reliance recently unveiled its latest venture, Reliance Intelligence, which will use Google Cloud infrastructure running on TPUs "With Ironwood, we can scale up to 9,216 chips in a superpod linked with breakthrough Inter-Chip Interconnect (ICI) networking at 9.6 Tb/s," said Google in the announcement.

Related Topics: #Ironwood TPU #Google Cloud #AI workloads #Gemini #Imagen #Veo #Anthropic #Claude #Inter-Chip Interconnect

Will the promised speed‑ups hold up in production? Google says Ironwood, its seventh‑generation TPU, will be generally available on Cloud within weeks, opening TPU v7 to a broader set of AI workloads. The chip claims a ten‑fold peak performance jump over TPU v5 and four times the performance per chip for both training and inference compared with TPU v6.

Customers can now tap that hardware for their own models, while Google continues to run Gemini, Imagen, Veo and other internal families on the same silicon. Yet the article doesn’t provide data on actual cost efficiency or how the gains scale across different workloads. Moreover, large‑scale Cloud users have previously adopted TPUs, but whether Ironwood will attract new segments remains uncertain.

In practice, the impact will depend on how developers integrate the hardware into existing pipelines and whether the performance claims translate into measurable productivity. The rollout is imminent, but concrete results are still pending.

Further Reading

Related Articles

Common Questions Answered

When will Google's Ironwood TPU be generally available on Google Cloud?

Google announced that the Ironwood TPU will be generally available on Cloud within the coming weeks. This rollout will let developers and enterprises provision the hardware without needing to join a lengthy waitlist.

How does the performance of Ironwood TPU (TPU v7) compare to earlier TPU generations?

The Ironwood TPU, also known as TPU v7, claims a ten‑fold peak performance increase over TPU v5. It also delivers roughly four times the performance per chip for both training and inference when compared with TPU v6.

Which internal Google AI model families are run on the Ironwood TPU?

Google uses the Ironwood TPU to train and deploy its Gemini, Imagen, and Veo model families, among other internal AI models. These workloads benefit from the chip’s higher compute density and speed.

Which external partner has historically used Google Cloud TPUs and recently expanded its partnership?

Anthropic, the company behind the Claude series of AI models, has long utilized Google Cloud TPUs for its workloads. The partnership has recently been expanded, allowing Anthropic to leverage the new Ironwood TPU for its next‑generation models.