模型

GB300

Q: GB300 最近有什么新动态？

traeai 已收录 2 篇与 GB300 相关的内容。最新一篇是「Training Kimi K2 and Qwen3 30B-scale models efficiently requires more than standard data-parallel tr...」，由 NVIDIA AI(@NVIDIAAI) 发布。

驱动 DGX Station 的芯片型号。

已跟踪 2 条高相关材料

TraeAI 观察

如果只读 3 篇

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

NVIDIA AI(@NVIDIAAI) · 7.2 分

NVIDIA Megatron Core新增对Muon、MOP、REKLS等高阶优化器的端到端支持，突破传统数据并行限制，显著提升Kimi K2与Qwen3等300亿参数模型在GB300/NVL72系统上的训练效率。

NVIDIA DGX Station 开始交付开发者和研究人员

AI HOT 精选 · 6.5 分

NVIDIA DGX Station 开始向开发者和研究人员交付，企业团队可通过 GB300 系统实现本地数据中心级性能。

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

NVIDIA AI(@NVIDIAAI)5月6日328 字 (约 2 分钟)

NVIDIA Megatron Core now offers end-to-end support for advanced optimizers like Muon, MOP, and REKLS, overcoming limitations of standard data parallelism to significantly accelerate training of 30B-scale models such as Kimi K2 and Qwen3 on GB300 and NVL72 systems.

入选理由：传统数据并行已不足以高效训练30B+大模型，需引入高阶优化器。

FeaturedTweet#NVIDIA Megatron Core#Muon#Qwen3#Kimi K2#LLM Training Optimization英文

NVIDIA DGX Station systems are beginning to arrive at the desks of developers and researchers.

AI HOT 精选6月2日105 字 (约 1 分钟)

NVIDIA DGX Station is being delivered to developers and researchers, enabling enterprise teams to achieve local data center-level performance with GB300-powered systems.

入选理由：NVIDIA DGX Station 已开始交付给开发者和研究人员。

FeaturedArticle#NVIDIA#DGX Station#Hardware英文

跨材料问答 · GB300

回答基于：GB300 相关 2 条材料