Zhiyuan FlagOS has completed Day 0 adaptation of DeepSeek-V4 for eight chips, covering Ascend, Hygon, Moorespeed, and others.

robot
Abstract generation in progress

Today, DeepSeek released the DeepSeek-V4-Pro 1.6T flagship model (1.86 trillion parameters) and the DeepSeek-V4-Flash 284B efficient model (284 billion). Led by the Institute of Intelligence Science, Zhiyuan FlagOS immediately performed full-scale adaptation of the two “giant” models, completing full adaptation and inference deployment of DeepSeek-V4-Flash across more than 8 AI chips, including Hygon, Muxi, Huawei Ascend, Moore Threads (FP8), Kunlun Chip, Pingtouge Zhenwu, Tianshu, and NVIDIA (FP8). FlagOS is also advancing the migration and adaptation of the DeepSeek-V4-Pro model to multiple chips, with an open-source release forthcoming. (Science and Technology Innovation Board Daily)

View Original
This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.
  • Reward
  • Comment
  • Repost
  • Share
Comment
Add a comment
Add a comment
No comments
  • Pin