Unverified50% confidenceFactExact time
The actual extent of cache savings depends on the hit rate; if 80% of input hits the cache, the actual input cost is approximately 0.8 × cache price + 0.2 × regular price.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCached 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.79% 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%).75% similarUnverifiedFor one-off inference where inputs never repeat or for batches of unique documents, because of the cache-write premium, enabling caching costs more with zero benefit, so it should simply be skipped.74% similarVerifiedThe cache read mechanism retains the KV cache for recurring long prompt prefixes, charging at an extremely low rate upon a cache hit; the leaked cache read price for Opus 5.5 is approximately 1/20 of the input price.73% similarUnverifiedFor applications that heavily rely on context reuse, the $0.10 cached read price is the primary source of cost savings.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1063622API
curl https://kongchang.com/api/v1/knowledge/claims/1063622MCP
get_claim(id=1063622)