Google has introduced TPU 8tx and TPU 8i Max, two additions to its eighth-generation TPU line.
TPU 8tx was designed with AMD, is based on TPU 8t, and is intended for AI model training and operation.
Google says it can isolate sensitive data and model-defining information while models are in use.
TPU 8i Max combines two TPU 8i chips in one unit.
TPU 8tx protects data during use
Google designed the TPU 8tx with AMD, building on the TPU 8t. The chip is intended for securely training and running artificial intelligence models.
According to Google, it protects and isolates sensitive data, as well as information that determines how models work, even while they are in use.
The company also says the chip can perform reasoning tasks and train models in parallel. This, Google says, could allow smaller models to be created and trained while the chip is running. One potential use is to train a model with corporate data, without having to transfer the data from one device to another.
TPU 8i Max doubles key features
The TPU 8i Max combines two TPU 8i chips into one unit. Google introduced the TPU 8i in April 2026. According to the company, the new model doubles the memory, processing power, and data transfer speed compared to two TPU 8i.
↳
READ ALSO
HACKS.GR / RELATED
AI Unlocked
Meta's Muse is now available on Apple Watch and Wear OS watches
↗
Google designed the TPU 8i Max to help models consider more information when answering queries.
Other AI companies and labs, including Microsoft, Amazon, OpenAI and Anthropic, are also developing their own chips and related programs, aiming to be less dependent on Nvidia, which the source describes as dominant.
>Stay connected with Hacks.gr
Set us as a preferred source on Google and find us on social media.
Preferred source
↗
end of article
Top comments (0)