Skip to content
Tagged articles

HyperParallel-FSDP

1 articles · Page 1 of 1
SmartApe AI
SmartApe AI
Nov 20, 2026 · Artificial Intelligence

China Telecom Trains 285B MoE Model on 8192 Ascend 910C Chips, Tops Hugging Face

China Telecom's Xingchen AGI Lab, with Huawei and HIT Shenzhen, pre-trained a 285B-parameter MoE model using 8192 Ascend 910C chips, detailing architecture optimizations (MLA, DSA, IndexShare, DSpark), system-level improvements (HyperParallel-FSDP, Owner-based Muon, selective recomputation), and the TeleGym RL framework for agent tasks, with a trillion-parameter model underway.

Ascend 910CChina TelecomHyperParallel-FSDP
0 likes · 13 min read
China Telecom Trains 285B MoE Model on 8192 Ascend 910C Chips, Tops Hugging Face