inference/research
verifiedhigh · 0.88updated 2026-10-03

Atom: Low-bit Quantization for Efficient and Accurate LLM Serving (Zhao et al.)

paper

Fields

id
sources/qlab-atom-paper
Publisher
arXiv
Kind
paper
Published
2023-10-29
Accessed
2026-10-03
notes
Mixed-precision block-outlier + fine-grained group quantization for W4A4 serving on A800.

Notes

Mixed-precision block-outlier + fine-grained group quantization for W4A4 serving on A800.

Cross-references

This record references no other records.