Advertisement
Semiconductors
TechBig Tech

Alibaba’s T-Head unit unveils details of AI chip designed to rival Nvidia’s GPUs

The launch of the Zhenwu 810E marks the latest initiative by China’s Big Tech firms to develop a domestic alternative to Nvidia’s AI chips

2-MIN READ2-MIN
Listen
T-Head, the semiconductor design arm of Alibaba, promotes its Zhenwu 810E chip as capable for both AI training and inference. Photo: Handout
Ann Caoin Shanghai
Alibaba Group Holding’s semiconductor design arm, T-Head, on Thursday unveiled details of its advanced artificial intelligence chip with capabilities said to be on par with Nvidia’s H20, days after the unit was reportedly being prepared for a potential spin-off and public listing.

According to information on T-Head’s website, the Zhenwu 810E was a “fully self-developed” parallel processing unit (PPU), an application-specific integrated circuit designed for both AI training and inference. Alibaba owns the SCMP.

Developed by T-Head to handle the massive data throughput required for generative AI systems, the Zhenwu 810E supports 96 gigabytes of High Bandwidth Memory 2 Enhanced – a 3D-stacked synchronous dynamic random-access memory standard designed for AI, machine learning and high-performance computing.
A single card with the T-Head PPU handles a bandwidth of 700 gigabytes per second, which made it on par with Nvidia’s H20 – a graphics processing unit (GPU) tailored by Nvidia specifically for China to comply with US export restrictions.
Select Voice
Select Speed
1x
AI-generated voice