Unverified50% confidenceSolutionExact time
After a long interruption (cache expiry), compaction is actually cheaper than simply continuing the conversation, because each subsequent turn then reads only about 15k tokens instead of the entire context.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedIn the session assumed by the article, a gateway labeled "50% off" but with cache reads at only a 20% discount and cache writes at full price actually gives a 28% discount, costing $12.13.73% similarUnverifiedThe reduction in compressed input tokens (49.5%) is greater than the reduction in total cost (35.6%), because output tokens, caching, and fixed overhead do not decrease proportionally with input compression.72% similarUnverifiedToken billing is calculated based on the total volume of input and output; the longer a conversation, the more each call pays for an increasingly bloated history.71% similarUnverifiedCompared to per-token pricing, per-minute pricing may result in higher costs for high-frequency short-utterance scenarios such as customer service conversations.68% similarUnverified提升分词器压缩率意味着在相同 token 预算内可以容纳更多原始文本，直接降低 API 调用费用68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1023333API
curl https://kongchang.com/api/v1/knowledge/claims/1023333MCP
get_claim(id=1023333)