Open-weight & owned stack · Quantization efficiency

Studies Explore Transform Coding For KV-Cache Compression

Sep 13, 2026

Two papers examined transform coding and rotated binary quantization as methods for compressing key-value caches during long-context LLM inference.

Read the original at awesomepapers.ioOpens the publisher's site in a new tab

Also covering this

KV Cache Compression Through the Lens of Transform Coding — Large Language Modelsawesomepapers.io, Sep 13

More in Open-weight & owned stack

Z.ai publishes GLM-5.3 open weightsSep 8 MiniMax H3 highlighted as open-weight multimodal modelSep 7 InternLM releases 397B multimodal Intern-S2 modelSep 13 DeepSeek Model Nearly Matches GPT-6 Astra DesignSep 10 DeepSeek presents V4 FlashSep 8