Huawei releases Flex:ai AI container technology to enable a single card to support multiple AI workloads at the same time.

Huawei released Flex:ai AI container software today. Through the technology of splitting computing power, a single GPU/NPU computing card can be divided into multiple virtual computing units, with a precision of up to 10%. This allows a single card to simultaneously handle multiple AI workloads, and can aggregate the idle XPU computing power of nodes in a cluster to form a "shared computing power pool". It is reported that Flex:ai will be open-sourced in the MoEngine community after its release.