Expired80% confidenceFactTime unknown
AgentMemory consumes only approximately 1,900 tokens per session, saving 92% compared to manually pasting context.
1
Sources
80%
Confidence
Medium-term (~90 days)
Relevance
6/3/2026
First Seen
Valid until: 9/1/2026(expired)
Sources
Related Entities
Related Claims
VerifiedAnthropic's Prompt Caching hit price is only 10% of the normal input price, meaning 90% of token costs can be saved.64% similarUnverifiedCache-read tokens account for more than 90% of overall token activity.63% similarUnverified在1.78百万token的长会话实测中，因为97%的token被缓存，发送token最多的配置反而最便宜63% similarUnverifiedInput tokens now account for more than 90% of total non-cached tokens, and the cost share of input context has risen from 47.5% to nearly 70%.63% similarUnverifiedAnthropic's Prompt Caching requires that prefix content be byte-for-byte identical and meet a minimum length threshold of approximately 1,024 tokens, delivering approximately a 90% token discount.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/39398API
curl https://kongchang.com/api/v1/knowledge/claims/39398MCP
get_claim(id=39398)