Whether an effective model configuration declares a byte-prefix prompt cache
(LLM.PrefixPromptCache === true). Absent, null or false all mean a block cache. Read through
ResolveEffectiveModelConfiguration (or AIEngineBase.GetEffectiveModelConfiguration) so
the most specific layer's answer wins.
Whether an effective model configuration declares a byte-prefix prompt cache (
LLM.PrefixPromptCache === true). Absent,nullorfalseall mean a block cache. Read through ResolveEffectiveModelConfiguration (orAIEngineBase.GetEffectiveModelConfiguration) so the most specific layer's answer wins.