Unverified50% confidenceFactExact time
The cache TTL for the subscription version can stay warm for one hour, while the API call mode defaults to expiring after five minutes.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/2/2026
First Seen
Valid until: 12/1/2026
Sources
Related Entities
Related Claims
UnverifiedAnthropic's Prompt Caching API requires the cached prefix to reach at least 1,024 tokens, the cache TTL is typically 5 minutes, and it can be automatically renewed during active sessions.71% similarUnverifiedL1缓存命中延迟约为4-5个周期，而寄存器访问为零延迟68% similarUnverifiedTime to First Token (TTFT) refers to the time interval between a user sending a request and the model outputting the first token, and is primarily constrained by the duration of the prefill stage.68% similarUnverified中心化API（如OpenAI）的首Token延迟（TTFT）通常在200-500毫秒之间67% similarUnverified一次AI API调用的延迟包括网络往返时间（50-200ms）、服务端排队、模型推理首Token延迟（TTFT，通常500ms-3s）及逐Token流式输出时间67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/842323API
curl https://kongchang.com/api/v1/knowledge/claims/842323MCP
get_claim(id=842323)