PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度
Commits on main
fix severe vram leak regression in auto-round format packing
dd7811e Wenhua Cheng committed 1 year ago
add support for scheme FP8_STATIC to export llm_compressor format (#816)
1089004 Heng Guo committed 1 year ago
AutoRoundMLLM supports scheme and fix device_map=dict regression (#801)
78fce45 Heng Guo committed 1 year ago