Huawei releases Flex: AI container technology enables single card to simultaneously carry multiple AI workloads

On November 21st, Huawei released Flex: AI container software, which uses computing power splitting technology to split a single GPU/NPU computing power card into multiple virtual computing power units with a precision of 10%. This enables a single card to simultaneously carry multiple AI workloads and aggregate idle XPU computing power from various nodes in the cluster to form a "shared computing power pool". It is reported that Flex: AI will be simultaneously open sourced in the Magic Engine community after its release.