Huawei MindSpore 1.2: The first domestic AI computing framework that supports training of large models

During the event of HDC.Cloud 2021, Huawei officially released the MindSpore 1.2, it is the first domestic AI computing framework that supports large model training with hundreds of billions of parameters. The latest version 1.2 brings the AI ​​framework field “automatic parallel, full-scenario AI, interpretable recommendation model”.

Huawei claims that MindSpore is the industry’s first fully automatic parallel framework based on network topology and cluster resource automatic sensing, and has developed the industry’s first 200 billion Chinese pre-training model based on fully automatic parallel capabilities.

MindSpore has realized the rapid application, efficient operation and effective collaboration of hardware devices in different scenarios of cloud, edge, and end. Through the full-scenario AI capability, the recognition rate of Huawei Watch GT’s wrist lift has been increased by 80%, the delay is less than 5ms, and the model is less than 1KB.

Back in March 2020, Huawei’s MindSpore Chief Scientist Chen Lei announced at the meeting that Huawei’s full-scenario AI computing framework MindSpore was officially open source. The company’s official data shows that since the open-source, the MindSpore community has more than 170,000 developers, with more than 240,000 software downloads, and it is used in more than 10 industries.

(Via)