Unverified50% confidenceFactExact time
The API returns content with an accurate token_count field based on tiktoken.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/12/2026
First Seen
Valid until: 12/11/2026
Sources
Related Entities
Related Claims
UnverifiedWhether the cache has actually been hit can be verified through fields such as cache_hit_tokens returned by the API.63% similarUnverifiedDevelopers who call the API directly receive precise token consumption data for each API call and can monitor usage in real time through the Dashboard61% similarUnverifiedtoken消耗的精确计量依赖于对Cursor官方API响应体中usage字段（包含prompt_tokens、completion_tokens）的实时解析59% similarUnverified语义缓存（Semantic Cache）通过计算新查询与历史查询的向量相似度，超过阈值时直接返回缓存答案，GPTCache等开源工具实现了这一机制，通常与TTL机制结合58% similarUnverifiedAnthropic需要在消息中添加cache_control字段来标记缓存断点，告诉API哪些内容应被缓存57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/912223API
curl https://kongchang.com/api/v1/knowledge/claims/912223MCP
get_claim(id=912223)