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

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

23 November 2025

Alay Dilipbhai Shah

Ben Athiwaratkun

Shuaiwen Leon Song

ArXiv (abs)PDF HTML Github

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

0 / 0 papers shown

Title
No papers found