// ◈ Quantization, Pruning & Distillation
AWQ (Activation-aware Weight Quantization)
Protects the ~1% most salient weights identified by activation magnitude.
AWQ 激活感知量化 2023 Advanced core
IN PLAIN WORDS
Weights tied to large-activation channels matter most. AWQ rescales those channels to reduce quantization error without backprop, ideal for edge deployment.
Used in: AWQ models in vLLM and TensorRT-LLM
01Original paper
AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
Ji Lin, Jiaming Tang, Haotian Tang, et al. · 2023
arXiv ↗ ✓ Verified · arXiv API
02Where it sits
Learn the left column first; the right column builds on this one.
Learn first
Then you can learn
- Nothing in the atlas builds on it yet
Related:SmoothQuant
Left: prerequisites · right: builds on this · dashed: related. Full map →
03Learning resources
Official docs, explainers, courses and reference code. All links checked.
04Where the course teaches it
On learning paths:AI Application Engineer
For AI / data
- Verification: verified, checked 2026-10-21.
- JSON: /api/learn/algorithms.json (id:
awq)