SOURCE-LINKED INTELLIGENCE

Minima-KV: Retention-Preserving KV Cache Compression with Mixed-Format Paged Attention

arXiv · AI, language, vision and robotics · article · Aug 24, 2026 · UTC

The key-value (KV) cache is a primary capacity and bandwidth bottleneck in long-context LLM serving. We present Minima-KV, a retention-preserving hierarchy for mixed-format paged attention. Recent and protected Anchor pages remain in FP8, while older non-anchor pages move to packed TQ3; every live-request page remains addressable. Format-specific kernels compute partial attention states and combine them through a globally normalized online-softmax merge, enabling direct heterogeneous decode without a cache-sized dense shadow. Across separate, configuration-bound Qwen3.6-27B profiles on a singl

Read original source ↗ Open in workspace

recordType

paper

region

Global