1
Full-Stack FP4: Stable LLM Pretraining with Quantized Projections, Optimizers, and Attention
首次实现全栈FP4量化预训练，突破LLM训练速度和显存瓶颈
arXiv:2607.04422v1 Announce Type: cross Abstract: Recent NVFP4 pretraining methods mainly target transformer linear layers, leaving optimizer states, …