OOM: Offload-Optimized Full-Parameter MoE Training via Memory-Execution Co-Design

arXiv:2611.11364 · cs.DC, cs.LG · Submitted 2026-11-08 · Read on arXiv

cs.DC, cs.LG

Submitted: 2026-11-08

Updated: 2026-11-08

Code: https://github.com/RegiaYoung/OOM-MoE

Terminology

Sources

Related papers