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

BitNet: Scaling 1-bit Transformers for Large Language Models (Ma et al., Microsoft)

paper

Fields

id
sources/qlab-bitnet-b1-paper
URL
https://arxiv.org/abs/2310.11453
Publisher
arXiv
Kind
paper
Published
2023-10-17
Accessed
2026-10-03
notes
The BitNet b1 (b1.0 / bit1) architecture paper with BitLinear, distinct from the b1.58 paper already in the repo (quant-bitnet-paper, 2402.17764).

Notes

The BitNet b1 (b1.0 / bit1) architecture paper with BitLinear, distinct from the b1.58 paper already in the repo (quant-bitnet-paper, 2402.17764).

Cross-references

This record references no other records.