In 2026, Tianjin Zhixin released the Tiangai 300 at the World Artificial Intelligence Conference. Based on the SIMT architecture, it supports scalar, vecto...
产品更新IT之家(RSS)
Today AI Intelligence Brief
In 2026, Tianjin Zhixin released the Tiangai 300 at the World Artificial Intelligence Conference.
Based on the SIMT architecture, it supports scalar, vector, and tensor computing. Its Attention efficiency exceeds 90%, and both the 64k context and MoE performance are 10% higher than the Hopper solution, with the first-token latency reduced by about 20%. This product has already been deeply adapted with domestic cloud providers and server manufacturers.
The readable text on this page was extracted from the public source and organized with attribution, publication time and the original link. Copyright remains with the original author and publisher.
In 2026, Tianjin Zhixin released the Tiangai 300 at the World Artificial Intelligence Conference. Based on the SIMT architecture, it supports scalar, vecto...
IT之家(RSS)2026-07-19T12:11:01.000Z
Scan to open this article
Aioga aggregates global AI updates and preserves source information for verification and citation.