#Huawei

'Huawei's CloudMatrix builds a peer-to-peer supernode combining 384 Ascend 910C NPUs and 192 Kunpeng CPUs to deliver high-throughput, low-latency LLM serving, with CloudMatrix-Infer optimizing MoE and KV cache workloads.'

Huawei has introduced Pangu Ultra MoE, a 718 billion parameter sparse language model optimized for Ascend NPUs using simulation-driven architecture and advanced system-level optimizations to achieve high efficiency and performance.