AWQ: Activation-aware Weight Quantization

Fields

id

sources/src5-awq-github

URL

https://github.com/casper-hansen/AutoAWQ

Publisher

AutoAWQ

Kind

repo

Published

2023-01-01

Accessed

2026-10-04

notes

AWQ provides activation-aware weight quantization for LLMs. Protects salient weight channels based on activation distribution. Supports INT4, INT8. Key for W4A16 quantization. Used by vLLM, TensorRT-LLM for quantized inference.

Notes

AWQ provides activation-aware weight quantization for LLMs. Protects salient weight channels based on activation distribution. Supports INT4, INT8. Key for W4A16 quantization. Used by vLLM, TensorRT-LLM for quantized inference.

Cross-references

This record references no other records.