DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads

Guangxuan Xiao, Jiaming Tang, Jingwei Zuo, junxian guo, Shang Yang, Haotian Tang, Yao Fu, Song Han

Keywords: Large Language Models; Long Context; Efficiency;

TL;DR: We significantly reduce both pre-filling and decoding memory and latency for long-context LLMs without sacrificing their long-context abilities.

Abstract:

Deploying long-context large language models (LLMs) is essential but poses significant computational and memory challenges.
Caching all Key and Value (KV) states across all attention heads consumes substantial memory.
Existing KV cache pruning methods either damage the long-context capabilities of LLMs or offer only limited efficiency improvements.
In this paper, we identify that only a fraction of attention heads, a.k.a, Retrieval Heads, are critical for processing long contexts and require full attention across all tokens.
In contrast, all other heads, which primarily focus on recent tokens and attention sinks—referred to as Streaming Heads—do not require full attention.
Based on this insight, we introduce DuoAttention, a framework that only applies a full KV cache to retrieval heads while using a light-weight, constant-length KV cache for streaming heads, which reduces both LLM's decoding and pre-filling memory and latency without compromising its long-context abilities.
DuoAttention uses a lightweight, optimization-based algorithm with synthetic data to identify retrieval heads accurately.
Our method significantly reduces long-context inference memory by up to 2.55$\times$ for MHA and 1.67$\times$ for GQA models while speeding up decoding by up to 2.18$\times$ and 1.50$\times$ and accelerating pre-filling by up to 1.73$\times$ and 1.63$\times$ for MHA and GQA models, respectively, with minimal accuracy loss compared to full attention.
Notably, combined with quantization, DuoAttention enables Llama-3-8B decoding with 3.33 million context length measured on a single A100 GPU. Code is provided in https://github.com/mit-han-lab/duo-attention.

Primary Area: foundation or frontier models, including LLMs

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 5112