Unverified50% confidenceFactExact time
Prompt caching typically has a short lifetime (on the order of minutes) and is keyed by provider, model, and content prefix.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedPrompt caching requires the request prefix to match byte for byte, and caches usually have a time to live (such as 5 minutes to 1 hour), after which they must be written again.76% similarUnverifiedPrefix caching is provided directly by model providers such as Anthropic and OpenAI, retaining the KV activation state of a static prompt prefix and reusing precomputed attention states, but it still incurs provider latency and network transmission overhead, with typical read latency of 200ms-1500ms.69% similarUnverified短期记忆通过在提示词中拼接历史对话实现，受限于模型上下文长度，通常4K到200K tokens不等68% similarUnverified对于单用户交互式场景推理速度达到5-10 token/s通常就能提供流畅体验，而代码生成、长文档摘要或多轮Agent调用往往需要30 token/s以上67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1055708API
curl https://kongchang.com/api/v1/knowledge/claims/1055708MCP
get_claim(id=1055708)