Unverified50% confidenceFactExact time
Anthropic's prompt caching is priced at $0.30 per million tokens for cache reads (about 10% of the standard input price), requires explicitly marking cache breakpoints, and carries a premium for cache writes.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedIn mainstream API pricing structures such as Anthropic's, the unit price of cache-read tokens is only 1/10 to 1/30 of standard input tokens.80% 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.78% similarUnverifiedPrompt caching allows vendors to cache the same repeated prefix in an Agent loop, with subsequent calls billed at the cache read price, in this example $0.20 versus $4 for input, 20 times cheaper.77% similarUnverifiedIn 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 $12.13.76% similarUnverifiedPrompt Cache refers to a provider caching already-processed identical prefix tokens, where the price of cache-hit tokens is typically only 10%–50% of the normal price.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1071958API
curl https://kongchang.com/api/v1/knowledge/claims/1071958MCP
get_claim(id=1071958)