SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models

On the reading lists

• #64 in Deep learning (2026 edition)

Why it matters

Enabled hardware-efficient W8A8 quantization by transferring activation outlier difficulty to weights.

Condensed from the expert lists' notes.

The record

First released

2022 (from arXiv 2211.10438)

Published

2023

Citations

2,022 citations, as counted by Papers With Code (CC BY-SA 4.0)

Named by

both expert lists

Score

0.308 · computed rank 587 · consensus 0.37, must-read 0, citations percentile 0.562 (2020s · deep-learning), awards 0, existing lists 0. Method