Why are cached input tokens cheaper with AI services?2 days agoAI pricingcost optimizationKV cachinghttps://xeiaso.net/notes/2026/why-llm-cached-token-cheaper/Copy Link