Unverified50% confidenceFactExact time
After the upgrade, higher cache hit rates are provided by default, allowing more input tokens to enjoy a cache input discount of up to 90%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/23/2026
First Seen
Valid until: 12/22/2026
Sources
Related Entities
Related Claims
UnverifiedSome platforms offer discounts of up to 75%–90% on input tokens that hit the cache.80% similarUnverifiedThe Prompt Caching mechanism can typically save 50%–90% of input costs.74% 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%.72% similarUnverified上下文缓存允许将已处理长文档前缀的KV缓存保存在服务端复用，部分提供商缓存命中的token折扣高达90%71% similarUnverifiedCache-read tokens account for more than 90% of overall token activity.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/943379API
curl https://kongchang.com/api/v1/knowledge/claims/943379MCP
get_claim(id=943379)