T
traeai
Sign in

产品

NVIDIA Megatron-Core

别名:megatron-core

NVIDIA的大型MoE训练框架

已跟踪 2 条高相关材料

TraeAI 观察

最近变化

2026-10-01 · Olmo-core 3采用DDP架构,使47B参数MoE模型吞吐量达52,000 tokens/s/GPU

为什么值得关注

NVIDIA Megatron-Core 被反复提及时,通常意味着它正在影响产品路线、开发者工作流或 AI 产业判断。这个页面把分散材料合并成一个可持续更新的观察入口。

Hugging FaceKimi K2LLM训练优化MoEMuon

相关材料

已收录 2 条与 NVIDIA Megatron-Core 相关的内容,按评分排序。

Hugging Face Blog 图标

Introducing Olmo-core 3: Open, scalable training infrastructure for large MoEs

Hugging Face Blog1481 字 (约 6 分钟)
85

Hugging Face发布Olmo-core 3,通过分布式训练优化将MoE模型参数容量提升至47B,吞吐量提升2.7倍。

入选理由:Olmo-core 3采用DDP架构,使47B参数MoE模型吞吐量达52,000 tokens/s/GPU

FeaturedArticle#MoE#分布式训练#Hugging Face#大模型英文
Training Kimi K2 and Qwen3 30B-scale models efficiently requires more than standard data-parallel tr...

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

跨材料问答 · NVIDIA Megatron-Core

回答基于:NVIDIA Megatron-Core 相关 2 条材料
    0 / 500

    AI may generate inaccurate information. Please verify important content.