KV Cache Reuse Efficiency on ShareGPT
30.5Match RateCachePrune
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CachePrunePrivacy setting=disabled, Reuse granularity=variable-length fine-grained, Model=Mistral-7B2026.05 | 30.5 | 177 | |
| EPICReuse granularity=fixed-size chunks, Chunk size=128 tokens, Model=Mistral-7B2026.05 | 8.57 | 209 | |
| CacheBlendReuse granularity=fixed-size chunks, Chunk size=128 tokens, Model=Mistral-7B2026.05 | 7.92 | 210 | |
| CacheCraftReuse granularity=fixed-size chunks, Chunk size=128 tokens, Model=Mistral-7B2026.05 | 7.73 | 213 |