Last released May 23, 2026
Semantic LLM answer cache — reuse paraphrased queries, cut latency and token cost.
Supported by