GPTQ vs AWQ vs SmoothQuant — key differences.
hardAnswer
- GPTQ (Frantar): layer-by-layer quantization with second-order approximation; fast + good accuracy for LLMs. AWQ (Lin): identifies salient weights via activation magnitudes; protects them from quantization → better preservation of critical channels.
- SmoothQuant (Xiao): migrates quantization difficulty from activations to weights via smoothing → makes activation-quantization tractable.
- Modern LLM serving mostly uses AWQ (best quality) + GPTQ + smoothquant depending on hardware.
- Runtime: vLLM / TensorRT-LLM support all.
Check yourself — multiple choice
- Same
- GPTQ: layer-by-layer 2nd-order approx (fast+good); AWQ: activation-magnitude salient weight protection (better quality); SmoothQuant: migrate quant difficulty act→weight via smoothing; AWQ modern best; vLLM/TRT-LLM support all
- Random
- Not real
GPTQ (2nd-order) / AWQ (salient protection) / SmoothQuant (act→weight); AWQ best quality.
#serving#llmops
Practise MLOps & Data Quality
215 interview questions in this topic.