← 返回列表

@NVIDIAAI: Training Kimi K2 and Qwen3 30B-scale models efficiently requires more than standard data-parallel tricks. NVIDIA Megatron Core now provides...

@NVIDIAAI 3 信息等级 3 1 噪音/剔除;2 较弱;3 普通事实;4 重要行业动态;5 极重大事件。该分数是信息显著性,不是投资建议。 发布:2026-05-04T21:00 抓取:2026-05-05 04:04
🔗 原文链接
摘要

NVIDIA Megatron Core 新增对 Muon、MOP 和 REKLS 等优化器的支持,旨在提升 GB300 GPU 和 NVL72 系统上训练 Kimi K2、Qwen3 30B 等大模型的效率。

客观事实
  • NVIDIA Megatron Core 支持 Muon 等高阶优化器
  • 针对 GB300 GPU 和 NVL72 系统优化训练效率
  • 用于训练 Kimi K2 和 Qwen3 30B 规模模型
NVIDIA Megatron Core GB300 GPU NVL72

原文

Training Kimi K2 and Qwen3 30B-scale models efficiently requires more than standard data-parallel tricks.

NVIDIA Megatron Core now provides end-to-end support for emerging higher-order optimizers like Muon, alongside research optimizers such as MOP and REKLS, to push training efficiency on GB300 GPUs and NVL72 systems.

Full breakdown 👇 https://t.co/D7E55OnCiK

likes: 79 | retweets: 12 | replies: 5 | views: 5792