Unverified50% confidenceFactExact time
According to a cross-vendor empirical comparison, reusing a 100,000-token prefix can reduce costs by 71% to 93%.
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 2024, the price per million input tokens for large model APIs fell approximately 90% compared to 2023, dropping from approximately $30 to $2–3.70% similarUnverifiedThe cost per 1 million output tokens declined by more than 90% between 2023 and 2025.68% similarUnverifiedAnthropic officially states that about 90% of Haiku 4.5 requests are below 100,000 tokens, and that average operating cost is about 75% lower.67% similarUnverified$27.5 represents a cost reduction of more than 80% compared to the original solution's $144.67% similarUnverifiedBased on Anthropic's official pricing, cache-hit tokens are typically billed at approximately one-tenth the cost of normal input tokens.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1058478API
curl https://kongchang.com/api/v1/knowledge/claims/1058478MCP
get_claim(id=1058478)