🧰 LMCache

The items that name it 4

• SARIA: Persistence-Aware KV Caching with Coordinated DRAM--SSD Reload for LLM Agents
  paperas classification.key entities2026-10-22
• Engrams Embedding Entendre: Codesign for Efficient DRAM/SSD Offloading
  articleas frameworks and tools2026-09-18

  Named as an ML community project supporting the benchmark.

• Vera Rubin NVL72 Agentic Inference: 67x better Performance per Dollar
  articleas frameworks and tools2026-09-14

  Named as an ML community project supporting the AgentX benchmark.

• vLLM v0.30.0
  repoas frameworks and tools

  KV cache caching system; cache_salt is validated before reaching it to prevent engine crashes