Skip to content

Commit 22c6fd8

Browse files
committed
update readme
1 parent aa3bcfb commit 22c6fd8

2 files changed

Lines changed: 11 additions & 3 deletions

File tree

README.md

Lines changed: 11 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -1,11 +1,19 @@
1-
# [HPCA 2026] BitDecoding
1+
![overview](imgs/title.png)
2+
3+
<div align="center">
4+
5+
## Efficient low-bit KV cache decoding
6+
27
[![arXiv](https://img.shields.io/badge/arXiv-2410.13276-b31b1b.svg)](https://arxiv.org/abs/2503.18773)
38
[![License](https://img.shields.io/badge/License-MIT-green.svg)](LICENSE)
49

10+
</div>
11+
12+
513
BitDecoding is a high-performance, GPU-optimized system
614
designed to accelerate long-context LLMs decoding with a low-bit KV
7-
cache. Achieve **3-9x speedup** than Flash Attention-v2.
8-
![overview](imgs/overview.png)
15+
cache. Achieve **3-9x speedup** than Flash-Decoding-v2.
16+
917

1018
## News
1119
* [2025.11] 🔥 BitDecoding has been accepted to HPCA 2025!

imgs/title.png

1.27 MB
Loading

0 commit comments

Comments
 (0)