KV-cache compression

VQKV

VQKV: High-Fidelity and High-Ratio Cache Compression via Vector-Quantization

On the frontierfirst seen Mar 17, 2026

Current frontier — recent, not yet superseded in the knowledge base

0 papers critique it · 0 beat it on benchmarks

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.