Skip to content

DEV Community

LLM Quantization Series' Articles

Back to Susheem Koul's Series

Cover image for How 8-bit quantization shrinks an LLM to half of its size — and why a single outlier weight can quietly ruin it

Oct 11

How 8-bit quantization shrinks an LLM to half of its size — and why a single outlier weight can quietly ruin it

#ai #llm #machinelearning #python

11 min read

Cover image for Training for quantization: QAT, 1-bit BitNet and 1.58-bit weights

Oct 17

Training for quantization: QAT, 1-bit BitNet and 1.58-bit weights

#machinelearning #llm #ai #deeplearning

8 min read