Corerain Technologies releases world’s first data flow AI chip CAISA

On June 23, Corerain Technologies held a product launch conference in Shenzhen, and released the world’s first data flow AI chip CAISA, positioning high-performance AI inference.
The chip has completed mass production.

On June 23, Corerain Technologies held a product launch conference in Shenzhen, and released the world’s first data flow AI chip CAISA, positioning high-performance AI inference.

The chip has completed mass production.

Corerain Technologies has achieved a technological breakthrough in the actual computing power of the chip through its independently developed data flow technology, which has increased the chip utilization rate by 10 times compared with similar products.

Third-party test data shows that with only 1/3 of the peak computing power, a CAISA chip can achieve up to 3.91 times the measured performance of similar products of NVIDIA.

Corerain Technologies’ data flow technology does not rely on a larger chip area and process technology, but uses data flow to control the calculation sequence to improve the measured performance and provide users with a higher price-performance ratio.

Since its establishment in 2016, Corerain Technologies has completed angel round, round A and round A+ financing.

It has R&D centers in Shenzhen, Shandong and London.

In 2018, it established the Artificial Intelligence Innovation and Application Research Institute, which is positioned to establish an artificial intelligence industrialization technology platform to support the rapid and practical implementation of the latest artificial intelligence technologies in various vertical fields.

It also initiated plans to carry out artificial intelligence courses training and scientific research cooperation with universities.

Disclaimer: This is an article created by Michael Liang for Chinasdg.org. You can find the original article here: https://chinasdg.org/2020/06/23/corerain-technologies-releases-world-s-first-data-flow-ai-chip-caisa/.

Leave a Reply

Please Login to Comment