Huawei

Huawei Finally Unveils Full Process of Near-Trillion MoE Ascend Training System

Huawei's Ascend + Pangu Ultra MoE achieves GPU-free, China-made trillion-parameter AI training with 41% MFU & 35K tokens/s throughput

Jun 1 •

Meng Li

Huawei NPU Runs Quasi-Trillion-Parameter Large Model

Huawei NPU achieves near-trillion-parameter AI model training, surpassing NVIDIA with Ascend-powered breakthroughs in efficiency & scalability.

May 9 •

Meng Li

#nojs-banner { position: fixed; bottom: 0; left: 0; padding: 16px 16px 16px 32px; width: 100%; box-sizing: border-box; background: red; color: white; font-family: -apple-system, "Segoe UI", Roboto, Helvetica, Arial, sans-serif, "Apple Color Emoji", "Segoe UI Emoji", "Segoe UI Symbol"; font-size: 13px; line-height: 13px; } #nojs-banner a { color: inherit; text-decoration: underline; } This site requires JavaScript to run correctly. Please turn on JavaScript or unblock scripts

ChipPub

Huawei