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

SqueezeLLM: Dense-and-Sparse Quantization (Kim et al.)

paper

Fields

id
sources/qlab-squeezellm-paper
URL
https://arxiv.org/abs/2306.07629
Publisher
arXiv
Kind
paper
Published
2023-06-13
Accessed
2026-10-03
notes
v4. Berkeley Sky Computing. Sensitivity-based non-uniform bit allocation + dense-and-sparse outlier store; up to 3-bit.

Notes

v4. Berkeley Sky Computing. Sensitivity-based non-uniform bit allocation + dense-and-sparse outlier store; up to 3-bit.

Cross-references

This record references no other records.