Unverified50% confidenceFactExact time
In that 30-day usage, 97.9% of token consumption came from cache reads, cache writes accounted for 1.75%, output accounted for only 0.34%, and fresh input accounted for just 0.01%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedIn this audit sample, 66% of the cost came from cache reads, 99% of input-side tokens were cache hits, while output (including thinking) accounted for only 12%.74% similarUnverifiedIn this audit, 85% of cache write tokens used 1-hour prompt caching, whose write unit price is higher than 5-minute caching; but reverse testing showed that switching everything to 5-minute caching would actually cost $1,781 more (+22.6%).74% similarUnverifiedA Prompt Caching cache hit read costs only 0.1 times the base input price, while the initial write cost is 2 times the regular input price.71% similarUnverifiedCached input tokens typically enjoy a discount of 25%–50% off the original price, as there is no need to recompute embeddings and attention weights—only the precomputed KV vectors need to be loaded from cache.70% similarVerifiedThe cost of cache reads should, by industry convention, be approximately one-tenth that of regular input tokens.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1057480API
curl https://kongchang.com/api/v1/knowledge/claims/1057480MCP
get_claim(id=1057480)