/moe-training
用 DeepSpeed 或 HuggingFace 訓練 Mixture of Experts (MoE) 模型。在有限算力下訓練大規模模型，或實作 Mixtral 8x7B、DeepSeek-V3 這類稀疏架構時使用。
- 安裝次數
- 984
- 專案星數
- 1.4 萬 ★
- 授權
- MIT
- 專案更新
- 2026-10-11
怎麼裝
npx skills add https://github.com/Orchestra-Research/AI-Research-SKILLs --skill moe-training安裝前請先看過原文內容。中文說明是 AI 翻譯，不保證安全。
原文
Train Mixture of Experts (MoE) models using DeepSpeed or HuggingFace. Use when training large-scale models with limited compute (5× cost reduction vs dense models), implementing sparse architectures like Mixtral 8x7B or DeepSeek-V3, or scaling model capacity without proportional compute increase. Covers MoE architectures, routing mechanisms, load balancing, expert parallelism, and inference optimization.