paper-with-me

Papers

Kitty: Accurate and Efficient 2-bit KV Cache Quantization with Dynamic Channel-wise Precision Boost

2025-11-23 · Haojun Xia, Xiaoxia Wu, Jisen Li, Robert Wu, Junxiong Wang, Jue Wang, Chenxi Li, Aman Singhal, Alay Dilipbhai Shah, Alpay Ariyak, Donglin Zhuang, Zhongzhu Zhou, Ben Athiwaratkun, Zhen Zheng, Shuaiwen Leon Song arxiv

The KV cache is a dominant memory bottleneck for LLM inference. While 4-bit KV quantization preserves accuracy, 2-bit often degrades it, especially on long-context reasoning. We close this gap via an algorithm-system co-design for mixed-precision KV caching: Kitty. On the algorithm side, extensive experiments show that Dynamic Channel-wise Precision Boost -- which ranks Key-cache channels by sensitivity and keeps only a small fraction at higher precision -- maintains near-zero loss in accuracy drop while approaching 2-bit memory. The main challenge is handling dynamic 4-bit channel boosts while keeping the page layout coalesced and the dequantization uniform, with no scattered reads or hard-coded masks. Kitty addresses these issues by decompose each mixed-precision Key page into two tensors with unified 2-bit precision. Based on this, Kitty provides a page-centric KV layout, Triton-compatible page dequantization kernels, and a lightweight runtime pipeline that preserves coalescing and avoids divergence. Across seven tasks and two model families (Qwen3, LLaMA3), Kitty cuts KV memory by nearly 8x with negligible accuracy loss, enabling up to 8x larger batches and 2.1x-4.1x higher throughput under the same memory budget. We release the full implementation of Kitty at https://github.com/Summer-Summer/Kitty.

📄 PDF Abstract BibTeX arXiv:2511.18643

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MixKVQ: Query-Aware Mixed-Precision KV Cache Quantization for Long-Context Reasoning

2025-12-22 · Tao Zhang, Ziqian Zeng, Hao Peng, Huiping Zhuang 외 arxiv

Long Chain-of-Thought (CoT) reasoning has significantly advanced the capabilities of Large Language Models (LLMs), but this progress is accompanied by substantial memory and latency overhead from the extensive Key-Value …

SKVQ: Sliding-window Key and Value Cache Quantization for Large Language Models

2024-05-10 · Haojie Duanmu, Zhihang Yuan, Xiuhong Li, Jiangfei Duan 외

Large language models (LLMs) can now handle longer sequences of tokens, enabling complex tasks like book understanding and generating lengthy novels. However, the key-value (KV) cache required for LLMs consumes substanti…

GPUQuantization

Accurate KV Cache Quantization with Outlier Tokens Tracing

2025-05-16 · Yi Su, Yuechi Zhou, Quantong Qiu, Juntao Li 외

The impressive capabilities of Large Language Models (LLMs) come at the cost of substantial computational resources during deployment. While KV Cache can significantly reduce recomputation during inference, it also intro…

Quantization

SVDq: 1.25-bit and 410x Key Cache Compression for LLM Attention

2025-02-21 · Hong Yankun, Li Xing, Zhen Hui-Ling, Yu Xianzhi 외

For the efficient inference of Large Language Models (LLMs), the effective compression of key-value (KV) cache is essential. Three main types of KV cache compression techniques, namely sparsity, channel compression, and …

Quantization

Plug-and-Play 1.x-Bit KV Cache Quantization for Video Large Language Models

2025-03-20 · Keda Tao, Haoxuan You, Yang Sui, Can Qin 외

Video large language models (VideoLLMs) have demonstrated the capability to process longer video inputs and enable complex reasoning and analysis. However, due to the thousands of visual tokens from the video frames, key…

Quantization