{"version":"2.1.311","anchor":"cache-miss-cost-now-counts-recached-tokens","canonical_anchor":"cache-miss-cost-now-counts-recached-tokens","heading":"Cache-miss statistics now also count recached tokens","tier":"internal","area":"Telemetry","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.311\/e\/cache-miss-cost-now-counts-recached-tokens","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.311","markdown":"### Cache-miss statistics now also count recached tokens\n\nA request now counts as a cache miss when its uncached and recached tokens together pass the limit, not uncached tokens alone\n\n**What**\n\nClaude Code keeps cost and usage statistics that include cache misses. Caching lets earlier parts of a conversation be reused so they cost less, and a miss is a request that could not reuse enough of it. A request used to count as a cache miss only when its uncached tokens passed a limit. It now counts when its uncached tokens and its recached tokens added together pass that limit.\n\n**Why**\n\nThe cache-miss count and cache-miss cost in your statistics can be higher than in the previous version for the same work. Keep this in mind when you compare figures across versions.\n\n- Area: Telemetry\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5\n- Scope: individual\n- Heads-up: no"}