Inference & Serving · 2022

Inference Quantization (GPTQ/AWQ/SmoothQuant/bitsandbytes)

Elias Frantar, Dan Alistarh, Ji Lin, Song Han, Guangxuan Xiao, Tim Dettmers, Luke Zettlemoyer

Introduced GPTQ, a one-shot post-training weight quantization method that compresses large language model weights to about 3-4 bits using approximate second-order (Hessian) information, letting big models run on far less memory with little accuracy loss.

Editorial record

Plain-language summary

GPTQ quantizes a pretrained model's weights layer by layer after training, using an approximation of the layer's Hessian to decide how to round each weight so that the overall error is minimized, and updating the remaining weights to compensate as it goes. This brings models down to roughly 3-4 bits per weight in one pass without retraining, shrinking memory enough to fit very large models on a single GPU. Related methods such as SmoothQuant (which shifts activation outliers into weights) and AWQ (which protects the weights most important to activations) address the same goal of low-bit inference through different trade-offs.

Knowledge graph

Relationships

Antecedents

Source record

Provenance

Record ID
P-404
Record created
2026-07-13
Last reviewed
2026-07-14
Record version
2

Citation caveat: Citation metadata is approximate and marked unverified in the source dataset.