MiniMax (official)· @MiniMax_AI · X·· 2 小时前AI 评分63
AI 导读
MiniMax 感谢 NVIDIA NeMo 团队为 MiniMax-M3.1-Flash 提供 day-0 微调支持。引用内容提到 Space Bunny 此前已处理 200T+ tokens，现已开放权重，NeMo-AutoModel 支持 FSDP2、专家、流水线和上下文并行，可扩展到 128 GPUs，NVFP4 checkpoint 可直接加载，并附 GitHub 文档链接。
正文 · 原文
Thanks to the NVIDIA NeMo team for day-0 fine-tuning support for MiniMax-M3.1-Flash! 🐰
What magic will this 🐰 bring in the post train scenario?
You’ve probably already used it: Space Bunny pushed 200T+ tokens before anyone knew it was MiniMax-M3.1-Flash. Now it’s open weights 🔥 Thank you @MiniMax_AI for open sourcing it! Day-0 fine-tuning in NVIDIA NeMo-AutoModel ✅ FSDP2, expert, pipeline, and context parallelism, ✅scaling up to 128 GPUs, and the NVFP4 checkpoint loads directly. https://github.com/NVIDIA-NeMo/Automodel/blob/main/docs/model-coverage/vlm/MiniMaxAI/minimax-m3-1-flash.mdx在 X 查看被引用的帖子
来源：MiniMax (official) · x.com